dnotes

------
Wed May  9 13:54:04 BST 2007

   Dirk Kostewa:
   overblend for electron density?

-----

trying to run render without render in path:

rendering ray trace number: 0
rendering ray trace number: 1
rendering ray trace number: 2
rendering ray trace number: 3
done raytrace.
calling render...
calling display...
((safe_scheme_command) Error in proc: key:  system-error  args:  (execlp ~A (No such file or directory) (..... scm_catch done 
scm_catch returns: 20152

2)))
..... scm_catch done 
scm_catch returns: #<unspecified>

Xlib: unexpected async reply (sequence 0xb9f30)!
Xlib: unexpected async reply (sequence 0xb9f4a)!
Xlib: unexpected async reply (sequence 0xb9f4b)!
Xlib: unexpected async reply (sequence 0xba0b6)!
Xlib: unexpected async reply (sequence 0xba0f1)!
Xlib: unexpected async reply (sequence 0xba0f3)!
lt-coot-real: Fatal IO error 14 (Bad address) on X server :0.0.

Some deprecated features have been used.  Set the environment
variable GUILE_WARN_DEPRECATED to "detailed" and rerun the
program to get more information.  Set it to "no" to suppress
this message.
ObjectCache: Leaked 0002 refs to <P 31 2" Nuvw = ( 240, 240, 132)>
ObjectCache: Leaked 0002 refs to <P 31 2">
47.510u 14.504s 12:56.93 7.9%   0+0k 0+0io 160pf+0w


interesting!


-----
Wed May 16 09:27:21 BST 2007

	User pickable atoms:

	a function that registers interest in n atom selections and a
	function to call when it has them.  We test in
	graphics-info-defines whether such a function exists and has
	all the number of picks that it needs, if it does and it
	doesn't then collect them and pass them on when it has all
	it needs.  When the function has been called delete it from
	the queue/vector of such functions (umm, better do the
	deletion just before the function call so that if there is an
	error in the scm function we don't jump away from the function
	deletion part).

----
Thu May 17 12:44:56 BST 2007

   INFO::
	use gtk_widget_get_name(widget) to find out about what the
	name of the widget is.


-----
Fri Jun 15 13:09:53 BST 2007

   rajiv.chopra@novartis.com:

   In real space refinement, don't move the atoms with zero occ.

------
 
  INFO::

  old guile convert to string:
    std::string name_target = SCM_STRING_CHARS(name_target_scm);
   new guile convert to string:
    std::string name_target = scm_to_locale_string(name_target_scm);

   all guile convert from string:
      SCM mess = scm_makfrom0str(a_string.c_str());


------
Fri Jul  6 13:24:43 BST 2007

  compiling guile-1.8.1

prfx=$HOME/testg

./configure --prefix=$prfx LDFLAGS=-L$prfx/lib CFLAGS="-g -I$prfx/include"

 ./configure --prefix=/lmb/wear/emsley/testg LDFLAGS=-L/lmb/wear/emsley/testg/lib CFLAGS='-O -I/lmb/wear/emsley/testg/include'

i.e. NO DEBUGGING!


cycle ~/compile/g/guile-gtk-2.0> ./configure --prefix=/lmb/wear/emsley/guile-1.8.1  --without-gtkgl

make 
make install

-
in guile-gtk-2.0's gdk-glue.c add

#define PANGO_PIXELS_CEIL(d) (((int)(d) + 1023) >> 10)
#define PANGO_PIXELS_FLOOR(d) (((int)(d)) >> 10)

- 
also need goosh:
./configure --prefix=/lmb/wear/emsley/guile-1.8.1
make install
and guile-gui:
 ./configure --prefix=/lmb/wear/emsley/guile-1.8.1
make install
and net-http:
 ./configure --prefix=/lmb/wear/emsley/guile-1.8.1
make install


----
Mon Jul 16 13:09:11 BST 2007

  if the first and last non-blank characters read from the scripting
  window are not ( and ) then add them. So

  read-pdb "test.pdb"

  will "work".

-----

Here's a description of the PIR format:

http://www.cmbi.kun.nl/bioinf/tools/crab_pir.html

----
Tue Jul 17 13:11:20 BST 2007

	Files that end in .fcf or .cif should not *first* be read as
	MTZ files.  Do the sensible thing first.

----
Mon Jul 16 23:18:03 BST 2007

	in creating a GTK2 gl context:


> WARNING:: Can't enable stereo visual - falling back 
> INFO:: Hardware stereo widget opened successfully
> INFO:: switch to hardware_stereo_mode succeeded

So in gl_extras():

We need a "stereo_successful" flag set before we get to saying 
"Hardware stereo widget opened successfully"


-----
Wed Jul 18 09:09:16 BST 2007

	Hydrogens naming problem with shelx.

	Shelx calls hydrogens attached to ILE CG2 HG2A HG2B HG2C,
	whereas refmac names are HG21 HG22 HG33.

	There are lots of other problems like this.


----
Sat Jul 21 10:25:29 MDT 2007

	problem with libcootaneer lib dependency when building in gtk1
	version.

* Don't see it any more.
	
-----
Wed Aug  1 19:48:19 BST 2007

	How about an Env Var to specify where to read and write
	startup files (state, history etc) so that you can run coot
	cleanly from the Desktop.
    
     Good idea.  Bit tedious though?  

-----
Wed Aug  1 22:55:35 BST 2007

     Add Chiral Volume Outliers to  some validation graph.  Geometry?

     ie. make the blocks big and blue (c.f. missing atoms)

     Priority Low, but cute and maybe useful.

-----
Thu Aug  2 03:20:36 BST 2007

   add a test for overlap-ligands - quite tricky.

----
Thu Aug  2 14:43:06 BST 2007

	unexpanded symmetry labels have 0-indexed symm labels.  It
	should be 1-indexed - Steven Sheriff

-----
Thu Aug  2 14:47:07 BST 2007

   How about some flexibility in file name output?

     How difficult is it to create (or, maybe, How would one go about
     creating a) naming system for output PDB files that is consonant
     with my usage?  Currently COOT gives a default of
     input_file_name-coot-0.pdb.  I typically name my files
     model_0i.pdb for the file out of refinement (before ligand, hence
     the zero -- yes I do it too ;-) ) and my first output file as
     model_0o1.pdb.  If I do some "editing" outside of COOT, I might
     then create model_0i2.pdb and if I make some changes inside COOT,
     the output file would be model_0o2.pdb, etc.  After the next
     cycle of refinement, I would come in with model_1i.pdb, etc.

	-- Steven Sheriff



-----
Sat Aug  4 14:36:06 BST 2007

	how about adding assign-pir-sequence to the state file?
	Problem is we don't store raw input, just the decoded sequence.
	We need both.
	something should be std::pair<std::string, std::string (orig, seq)
	not just std::string

        Hmm... perhaps a simple class then, because we need a flag to
        say that this molecule has a sequence assigned to it.


------
Fri Aug 10 12:56:36 BST 2007

   add a side chain flip test.

   check that the OE1 was where the OE2 was
   and vica versa.

   also test torsions from running correct-nomenclature 


-----
Mon Aug 13 15:43:54 BST 2007

	Add a test

	Use multiple wiggle ligands. Delete stuff and do it again.

-----
Mon Aug 13 16:52:03 BST 2007

	fit SAH into 2an3.

	It's really slow to converge.  Why?


-------
Mon Aug 20 18:08:16 BST 2007

	Add lots of tests for the space groups of shelx .res files
	using show-spacegroup


----
Sun Aug  5 00:29:27 BST 2007

   Revisit the --as-needed patch from Donnie Berkholz

   Try to apply (bits of) it and see what happens - problems on
   feisty, I guess

---
Sun Aug  5 00:30:26 BST 2007

   The X11 libs dependencies need to be revisited - I had to install
   some stuff to compile - they should be tested for in configure.

   Try with a gutsy pre-release?

----
Sun Aug  5 00:32:03 BST 2007

   The Coot FAQ should be written as a texi (article?) document, so
   that it has an index and toc.

----
Sat Aug 11 13:15:35 BST 2007

   Icons
   http://jimmac.musichall.cz/icons.php

   Think about icons for info dialog and the like.

----
Sat Aug 11 15:51:42 BST 2007

    Open Clip Art Notes:

    world in connect-to-world is "cartoon globe" by lunanaut

    crispy computer mouse for scroll wheel.

-----

http://www.itbusiness.ca/it/client/en/home/News.asp?id=44529&cid=11
	
The Novell SUSE Linux subscription model is considerably less painful,
Chevalier argues, because it doesn't hit the capital expense budget
and it can often be considered an operational expense, which is better
from a tax perspective.


--------

Tue Aug 21 14:26:00 BST 2007

   Thoughts for a tutorial or some such.

   "Assessing structure quality with Coot"
    (in response to EJD's comments on CCP4bb 2hr0)

    Structure with 0 occupancy

	What is a good (bad) Ramachandran?
	    
            What is a bad Kleywegt plot?
	
	Good (bad) geometry graph.

	Chiral volume outliers

	Cis-peptides, planar peptide restraints

	Good density fit graph (and how to scale it)

	Nomenclature errors in rotamers - and how to fix them
	   Missing atoms

	How to use with Molprobity (web output and local exes)

	B-factor variance.  What's good and bad.
	
   Priority: Low


-----
Tue Aug 21 14:44:14 BST 2007

	Note to self:

	B factor average, variance vs resolution for each residue type.

	Probably best to have a shell script to provide the pdb files
	
	c++ program reads pdbs and does b-factor analysis, 

            on completion appends to files "ARG.tab" "ASP.tab" etc
            with average and variance for each residue of that type in
            that pdb file.  Also the resolution.

	    only do 1 chain of each pdb file.

            use Coot libs - make it a Coot program.


    Priority: Low


-------
Tue Aug 21 16:50:52 BST 2007

  ionic waters:

  http://biochem.wustl.edu/~enrico/

  e.g. WASP
  Reference: Nayal, M., & Di Cera, E. (1996) J. Mol. Biol. 256: 228-234

  Nayal M and Di Cera E. Valence screening of water in protein
  crystals reveals potential Na+ binding sites. J Mol Biol 256:
  228–234, 1996

  Looks straightforward.  Except symmetry must be used to find neighbours.

  GSoC project?


-----
Tue Aug 21 18:15:47 BST 2007

	blender's iksolver for inverse kinemetic (loop fitting).
	Source code not obviously in 2.44.

        But see this
   http://www.koders.com/cpp/fid45A6AC666E04B001C8A397CC9CEA3A46D801EB30.aspx?s=cdef%3aparser
        Hmm.


------
Fri Aug 24 15:40:17 BST 2007

electron_density_gradient_at_point() in simple-restraint.cc used
template type double.  Would things be faster with a float?

clipper::Grad_orth<double>
coot::restraints_container_t::electron_density_gradient_at_point(const clipper::Coord_orth &ao);


----
Thu Aug 30 10:15:37 BST 2007

  From reading JED mail

  Do some speed tests on regularizing with and without peptide plane restraints.


  Interface to change the weight of the peptide plane restraint

  Interface to turn on peptide plane restraints for refinement (or regularizing) only?

  Allow the display of anisotropic atoms per molecule - from the
  scripting layer at least.


-----
Fri Aug 31 16:28:42 BST 2007

	Do a NCS ghosts for a blue map, 

	Do it again for the difference map.

	Does it recalculate the NCS operators?

	   It shouldn't.


------
Tue Aug 28 14:31:25 BST 2007

   And if the display manager would automatically resize itself when
   more objects are loaded, would be great.
 
  Marko Hyvönen
  Department of Biochemistry    
  University of Cambridge        

   Priority: Low, Difficulty: Medium.
   A thing for GSoC.


-----
Tue Aug 28 14:48:50 BST 2007

      http://blog.mithis.net/
      lython and SchemePy 

------
From: 	Matthew Wilce <matthew.wilce@med.monash.edu.au>
To: 	emsley@ysbl.york.ac.uk
Subject: 	Coot and polynucleotide
Date: 	Tue, 17 May 2005 10:39:44 +1000  (01:39 BST)

    Functionality for both RNA and DNA

    the ability to change RNA into DNA and visa versa

    mutation of bases

    rotamer flipping (manual and automated)

    fragment library for each nucleotide

    a version of baton would be useful - either base to base and/or  
    phosphate to phosphate

      o is that still needed?

    insertion of nucleotides

    addition of a nucleotide at 3' or 5' end of polynucleotide

      o Done?

-----

   Venki Ramakrishnan:

	"CA mode" for nucleic acids should draw a P and C1' trace.

-----

   Write a FAQ for sucking of metal ligands into metal density.

-----

   make dist runs a scsh script.  configure should check for scsh?
   (not easy to convert to guile) 

 
-----
Sun Sep 16 00:33:44 BST 2007

	Suggestion on the coot list in the last week:

    o Better algorithm to find picked atoms (prefer atoms at the front).

* Done.

    o "Reverse" chi angle fragment tail/dog

* Done.

    o Cache libcheck results for faster "Get Monomer"

* Done.

-----
Sun Sep 16 01:25:09 BST 2007


texi2html ../../coot/doc/user-manual.texi
*** Undefined node of menu_entry (Dictionary): * Dictionary::
*** Undefined node of menu_entry (Planar Peptide Restraints): * Planar Peptide Restraints::
*** Undefined node of menu_entry (The <TT>UNK</TT> residue type): * The @t{UNK} residue type::

	Priority: Medium: Difficulty: Low

----
Sun Sep 16 12:31:03 BST 2007

	refine-zone, regularize-zone, auto-refine-zone all ignore the
	alt-conf specifier.  They should not.

	Priority: Medium.

------
Wed Sep 19 17:20:41 BST 2007

	doing a flexible ligand fit in a ligands without flexible
	bonds - e.g. SO4 gives non-optimal message ("Did you forget to
	read the dictionary?").

-----
Tue Sep 18 12:01:13 BST 2007

   Interesting/useful?
	OpenBabel:
	http://openbabel.sourceforge.net/wiki/Main_Page

        pygtk-openbabel - A PyGTK frontend to OpenBabel A PyGTK
        frontend to OpenBabel that utilizes the power of
        python. Included is a graphical frontend to PubChem

	PubChem:
        http://pubchem.ncbi.nlm.nih.gov/


	Priority: Medium: Difficulty: Low

----
Tue Sep 18 16:12:29 BST 2007

   Mischa 

	if pick centering on an atom fails, can you fall back to test
	against having clicked on the intermediate atoms?

	   Indeed.  Why not.


-----
Fri Sep 21 18:17:52 BST 2007

	
	When there is a sytax error in ~/.coot, is it obvious what has
	happened?


-----

        RSR when map is ambiguous cancels rsr.  It should not.

----
Wed Sep 26 13:12:24 BST 2007


   Regularize residue A 11, 2izq

   Crashes in recently touched (well, the replace type has been
   changed) replace_coords()

(gdb) where
#0  molecule_class_info_t::replace_coords (this=0x87bfc7c, asc=@0x866b518, change_altconf_occs_flag=false)
    at ../../coot/src/molecule-class-info.cc:4564
#1  0x080c4a9f in graphics_info_t::accept_moving_atoms (this=0xbf8b8463) at ../../coot/src/graphics-info.cc:1364
#2  0x0815d612 in accept_regularizement () at ../../coot/src/c-interface-build.cc:5093
#3  0x08218ae2 in on_accept_reject_refinement_accept_button_clicked (button=0x90a88e8, user_data=0x0)
    at ../../coot/src/callbacks.c:2283
#4  0xb6e009d9 in g_cclosure_marshal_VOID__VOID () from /usr/lib/libgobject-2.0.so.0
#5  0xb6df362b in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0
#6  0xb6e04103 in ?? () from /usr/lib/libgobject-2.0.so.0


----
Wed Sep 26 14:52:14 BST 2007

	read junk*.pdb in order (starting at junk.pdb) and it crashes on junk2.

	note that model 0 is not a valid model.


    
-------


	A168 on 1A0F fails to properly find the right rotamer, even
	though we select the rotamer by hand and do rigid body refine
	it works fine. 

	Bug?


-------

	Residue 13 of 1UBQ fails to Rigid body refine after chi1 was 
        rotated by 180.
	
	Bug?

        In fact, it looks like nothing with Rigid body refine.  Hmmm!

------

    What about interactions with PISA?

* Scheduled for 0.6.

------



 Peter's talk:

   need to set the ccp4 project from the start script.

- Done.  Needs testing.  Also, the correct ccp4 project should
  selected on creation of file-selection.  This may be a bit harder.

-----

    Frank:

    add test peptide flips cf a database

    This is quite important.

-----
Sat Sep 29 20:04:56 BST 2007

CKChuang@LBL.GOV

   Also, when I reverse direction of a chain, if I use Reverse
   Direction followed by Ca Zone-> Mainchain, a few CAs will remain
   without a chain.  If I delete the preexisting residues and try to
   Rotate/Translate them myself they are automatically reversed back
   to their original direction once I try to Regularize Zone.  Is
   there a reason for this or a way to overwrite or go around this?


----
Thu Sep 27 09:27:26 BST 2007

	Ramachandran plot from atom selection (no need for a gui for
	this just yet).

---
Thu Sep 27 09:30:20 BST 2007

	John Berrisford complains about OXT refinement.

	Make sure Coot properly refines OXT at some stage (by
	correctly handling modifications).

-----
Tue Oct  2 09:05:25 BST 2007

   A must-do for 1.0:

        ***** 

	Put backrub into minimizer for auto-fit-rotamer - as an
	alternative.

        ***** 
----
Wed Oct  3 12:23:09 BST 2007


   NCS Ligands

   Make the NCS ghost matrices available at the scripting layer

	(let ((rtop ((get-ncs-matrix imol "A" "B"))))

	   ;; now apply rtop to the ligand "C" in the "A" pocket to
	   ;; make "D" in the "B" pocket

----
Wed Oct  3 18:59:34 BST 2007

   write a triv app with latest glade with icons and libglade

----
Sat Oct  6 08:04:36 BST 2007


pkg-config libglade-2.0 --cflags

-I/usr/include/libglade-2.0 -I/usr/include/gtk-2.0 -I/usr/include/libxml2 -I/usr/lib/gtk-2.0/include -I/usr/include/atk-1.0 -I/usr/include/cairo -I/usr/include/pango-1.0 -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -I/usr/include/freetype2 -I/usr/include/libpng12  

pkg-config libglade-2.0 --libs

-lglade-2.0 -lgtk-x11-2.0 -lxml2 -lgdk-x11-2.0 -latk-1.0 -lgdk_pixbuf-2.0 -lm -lpangocairo-1.0 -lfontconfig -lXext -lXrender -lXinerama -lXi -lXrandr -lXcursor -lXfixes -lpango-1.0 -lcairo -lX11 -lgobject-2.0 -lgmodule-2.0 -ldl -lglib-2.0  


----
Tue Oct  9 22:19:02 BST 2007

    GSOC project?
	nice graphic output of SSM results.
hari jayaram <harijay@gmail.com>


-----
Thu Oct 11 23:20:47 BST 2007

	convert axes.scm to axes.py for Paolo.


----
Mon Oct 29 09:36:34 GMT 2007

   Clemens:

   Add function that does
   copy_residue_range_from_ncs_master_to_others(int imol, const char *master_chain_id, 
					  int residue_range_start, int residue_range_end)
   but to a specific chain only:

   i.e.:

   copy_residue_range_from_ncs_master_to_chain(int imol, const char *target_chain_id, 
					  int residue_range_start, int residue_range_end)

   which looks up the master_chain_id from the ghost of the target_chain_id
   

------

   I would like to report a bug in the 0.4 pre-release.  In the clash 
   panel generated by probe, the vertical bar doesn't work anymore.  
   Could you fix it?n

Jianghai


------


I don't need anything, but it said I should send this to you.

Thanks,
Adam Okerlund

----------------------------------------------

Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `/usr/local/coot/bin/coot-real'.
Program terminated with signal 11, Segmentation fault.
#0  0x081acbad in lookup_widget (widget=0x33203a66, 
    widget_name=0x84a151c "geometry_distance_togglebutton") at support.c:33
        in support.c
#0  0x081acbad in lookup_widget (widget=0x33203a66, 
    widget_name=0x84a151c "geometry_distance_togglebutton") at support.c:33
#1  0x080d31ec in graphics_info_t::unset_geometry_dialog_distance_togglebutton
    (this=0xffc0e7cb) at graphics-info.cc:5328
#2  0x0811fdda in graphics_info_t::check_if_in_geometry_range_defines (
    this=0xffc0e7cb, event=0x8662118) at graphics-info-defines.cc:393
#3  0x081232c6 in graphics_info_t::check_if_in_range_defines (this=0xffc0e7cb, 
    event=0x8662118, state=@0xffc0e7bc) at graphics-info-defines.cc:201
#4  0x081aab02 in glarea_button_press (widget=0xab91428, event=0x8662118)
    at globjects.cc:3344
#5  0x55f24f64 in gtk_marshal_BOOL__POINTER (object=0xab91428, 
    func=0x81aa812 , 
    func_data=0x0, args=0xffc0e938) at gtkmarshal.c:28
#6  0x55f594d2 in gtk_handlers_run (handlers=0xc4cc200, signal=0xffc0e8b8, 
    object=0xab91428, params=0xffc0e938, after=0) at gtksignal.c:1917
#7  0x55f5bba3 in gtk_signal_real_emit (object=) at gtksignal.c:1477
#8  0x55f5c1e8 in gtk_signal_emit (object=) at gtksignal.c:552
#9  0x55f975a1 in gtk_widget_event (widget=0xab91428, event=0x8662118)
    at gtkwidget.c:2864
#10 0x55f230b8 in gtk_propagate_event (widget=0xab91428, event=0x8662118)
    at gtkmain.c:1378
#11 0x55f241ee in gtk_main_do_event (event=0x8662118) at gtkmain.c:803
#12 0x55fdf71b in gdk_event_dispatch (source_data=0x0, 
    current_time=0xffc0ed50, user_data=0x0) at gdkevents.c:2139
#13 0x56012bc9 in g_main_dispatch (dispatch_time=0xffc0ed50) at gmain.c:656
#14 0x5601345d in g_main_iterate (block=) at gmain.c:877
#15 0x5601365f in g_main_run (loop=0x8a054e0) at gmain.c:935
#16 0x55f2463f in gtk_main () at gtkmain.c:524
#17 0x080d233b in c_inner_main (closure=0x0, argc=1, argv=0xffc0f144)
    at c-inner-main.c:217
#18 0x556258c0 in invoke_main_func () from /usr/local/coot/lib/libguile.so.12
#19 0x5562586d in scm_boot_guile_1 () from /usr/local/coot/lib/libguile.so.12
#20 0x5562556c in scm_boot_guile () from /usr/local/coot/lib/libguile.so.12
#21 0x080d2018 in c_wrapper_scm_boot_guile (argc=1, argv=0xffc0f144)
    at c-inner-main.c:237
#22 0x080cd46e in main (argc=) at main.cc:402
Current language:  auto; currently c


-----
Fri Oct 19 21:46:34 BST 2007

   Message from EJD.

   Check that the skeleton colour works correctly.

-----
Thu Oct 25 11:09:07 EDT 2007

reading in a minimal state file:

Bad view quaternion

----
Thu Oct 25 18:23:31 BST 2007

    Can we fit the helix at residue 122 in the CSHL Au Lysozyme
    dataset?

-----
Tue Oct 30 15:24:25 GMT 2007

crash from EJD, not much to report:


#4  0x089626e6 in gtk_canvas_re_unrealize (item=0x1b5ad798) at gtk-canvas-rect-ellipse.c:598
#5  0x0895a74d in gtk_canvas_group_unrealize (item=0x1a68e428) at gtk-canvas.c:1796
#6  0x0895bb39 in gtk_canvas_unrealize (widget=0x1b330028) at gtk-canvas.c:2568

    validation graph open, then molecule closed?

    validation graph open, and graph is updated?  residue deleted?
    residue generated (add terminal residue) then graph is asked to be
    updated?

    
----
Tue Oct 30 15:49:12 GMT 2007

   On starting up coot, if imol_refinement_map() is undefined, then
   set insensitive the map fitting functions (e.g. refine zone,
   autofit rotamer).

   When it is set, we can activate those buttons (and menu items?).


----
Tue Oct 30 15:52:39 GMT 2007

   CSHL CRASH:

   on calculating a skeleton after deleting a map (and skeleton) and
   reading in a new map.

   
----
Tue Oct 30 15:53:49 GMT 2007

    Remove probability limit when using Richardson Rotamers.

* Done.

----
Tue Oct 30 15:54:27 GMT 2007

    Editing chi angles on a reduced pdb file failed to rotate the
    hydrogens (maybe a dictionary/version 3.0 problem).

    Also (set-find-hydrogen-torsions 1) fails.  Same cause?

---
Mon Nov  5 11:49:30 GMT 2007

  Maria Solà Vilarrubias: 

   I want to display ILE, LEU, VAL residues with just sidechains (in a
   different colour), the rest with Mainchain CAs.


----
Tue Nov 13 23:30:57 GMT 2007

   what is read_number in cif reading for. 

   try to get rid of it.

   examine read_number_in usage here:

  coot::protein_geometry::init_refmac_mon_lib(std::string ciffilename, int read_number_in)


----
Tue Nov 13 23:42:21 GMT 2007

	coot --recover

    reads history file (history file needs to be written when commands
    executed - not at exit time) to get maps.

    reads most recent saved files in coot-backup.

    recover those written < 5 mins ago (or some such).


----
Tue Nov 13 23:30:57 GMT 2007

   Gregory C. Benison NMR program:
   http://oregonstate.edu/~benisong/software/burrow-owl/
   now at
   http://sourceforge.net/projects/burrow-owl

-----
Fri Nov 16 20:23:19 GMT 2007

   How do I change my residue to have insertion code and the following
   residues to have -1 offset and it all to be joined up?

   e.g.:
   (change-residue-number 0 "C" 86 "" 85 "A")
   Which is fine, execpt I can't then move C/85A to the A chain.


----
Sat Nov 17 12:14:46 GMT 2007

	phosphorylate-active-residue: 

	what about phospholysine?

----
Sun Nov 18 16:47:57 GMT 2007

	update Map Transformation example in user-manual


-----
Thu Nov 22 16:28:41 GMT 2007

EJD crash:

#0  0x0844eb3a in molecule_class_info_t::intelligent_this_residue_mmdb_atom ()
#1  0x0834be67 in graphics_info_t::on_go_to_atom_residue_tree_selection_changed_gtk1 ()
#2  0x0837f6c8 in on_go_to_atom_atom_list_selection_changed_gtk1 ()
#3  0x089d2c26 in gtk_marshal_NONE__POINTER_INT_INT ()
#4  0x08a0605b in gtk_handlers_run ()
#5  0x08a0520b in gtk_signal_real_emit ()
#6  0x08a02b07 in gtk_signal_emit ()
#7  0x089d2a80 in gtk_marshal_NONE ()
#8  0x08a05118 in gtk_signal_real_emit ()
#9  0x08a02b07 in gtk_signal_emit ()
#10 0x08a2dba0 in gtk_tree_select_item ()
#11 0x08a2c571 in gtk_tree_button_release ()
#12 0x0c520378 in ?? ()
#13 0x00000000 in ?? ()



------
texi2html ../../coot/doc/user-manual.texi
*** Undefined node of menu_entry (Dictionary): * Dictionary::
*** Undefined node of menu_entry (Planar Peptide Restraints): * Planar Peptide Restraints::
*** Undefined node of menu_entry (The <TT>UNK</TT> residue type): * The @t{UNK} residue type::
mkdir -p chapters ; \
cd chapters ; \
splitstr='-split_chapter' ; \
texi2html --version ; \
if [ $? = 0 ] ; then splitstr='-split=chapter' ; fi ; \
texi2html $splitstr ../../../coot/doc/user-manual.texi ; \
cd ..
1.64
*** Undefined node of menu_entry (Dictionary): * Dictionary::
*** Undefined node of menu_entry (Planar Peptide Restraints): * Planar Peptide Restraints::
*** Undefined node of menu_entry (The <TT>UNK</TT> residue type): * The @t{UNK} residue type::


----
Wed Dec  5 14:13:31 GMT 2007

    Plans for multi-stick representations

    class line_representation {
      representation_type { atom_bond, CA, CA+ligand etc }
      bonds_container
      displayed_flag
      bond_thickness
      colour_rotation
      atom_selection_description { residue_type, residue_range_type, chain_type
	chain_id 
        resno_1
        resno_2;
	insode_1;
	insode_2;
      }
    };
      
    We shall use a std::vector<line_representation> line_rep_vec;  
    New items are pushed back.  You can't delete an item from this
    vector, mearly not display it (the handles are passed back - and
    we don't want them to go out of date).

    Particular alt confs cannot be selected for.

    Questions:  How shall we create new represenations for this molecule?

    // return 
    int add_representation imol represenation-spec

    // used by GUI
    undisplay_represenation imol respresentation-number

    undisplay_represenation imol residue_spec
    which needs to be unambiguated. Test in this order:

     o Was there a line_representation that has that residue_spec?
     o Was there a residue_range line_representation that encompassed that residue_spec?
     o Was that residue in a chain that has a line_representation?


----

	Tested/Not Tested:

   green_box on rama plot, on edit backbone angles crashed because
   canvas items were not being generated properly (not assigned to a
   canvas item before being pushed back into vector).

-----
Wed Dec 12 13:20:33 GMT 2007

   Make nice images for nucleic acid base chooser.  
 	
   The buttons don't need to be small.


-----
Thu Dec 13 17:48:24 GMT 2007

   write a test for fix-nomenclature-errors

-----
Sat Dec 15 12:18:46 GMT 2007

   1OHY is good for displaying NCS skipping NCS jumping

   as is  2x2h

   79, 137

   as is 3vhn 3fold with some sugar

-----
Sat Dec 15 13:12:08 GMT 2007

	Randy Read: 

    (copy-residue-range-from-ncs-master-to-others 0 "A" 1 5) needs to
    work if residues 1 to 5 are not in the B and C chains (because
    I've just built them in the A chain).

* Done.

----
Sat Dec 15 13:13:45 GMT 2007

    Randy Read:

	I want a tool for applying symmetry operator to a fragment

    How about this then:

    Other Modelling Tools -> Apply Symmetry to fragment
    [click on a atom]

    Coot finds fragment by selecting residues that have contiguous residue 
    numbering (does minimol do that?)
 
    -> Intermediate atoms

    Now pop up a gui

    A box of symmetry ops:
    X, Y, Z,
    -X+1/2, -Y, Z+1/2 
    (etc)

       X:  - +
       Y:  - +
       Z:  - +

    + "I'm feeling lucky", which should try to find the best packing
      itself (somehow)

----
Mon Dec 17 12:50:36 GMT 2007

   save_stat_file_name() returns a char * - yeuch.  It should return a SCM
   and be renamend save_stat_file_name_scm().  Also save_stat_file_name_py().

   Priority: Low
   Difficulty: Trivial

* Done.

-----
Mon Dec 17 16:52:20 GMT 2007

   Steven Sheriff:

   On calculating blobs from a difference map:

10.44 (10.44 sigma) at xyz = (  22.62, 4.526, 70.71)
10.40 (10.40 sigma) at xyz = (  22.35, 3.493, 70.95)   #  1.06 A separated from one above;
10.31 (10.31 sigma) at xyz = (  22.6, 4.587, 70.69)  # # 0.07 A separated from two above; 1.15 A separated from one above

"My view is that I don't want to see all 3 of these in the list, one would do. "

    One can see something similar if you peak search the difference
    map, I think.  Tweak the difference map algorithm to filter out
    sub-sites that are close to already found sites.  
    Weighted average, maybe?


------
Tue Dec 18 13:41:58 GMT 2007

in 
void
graphics_info_t::execute_rotate_translate_ready() { // manual movement

   CAtom *atom1 = molecules[imol_rot_trans_object].atom_sel.atom_selection[rot_trans_atom_index_1];
   CAtom *atom2 = molecules[imol_rot_trans_object].atom_sel.atom_selection[rot_trans_atom_index_2];

Which can be a crash of atom1 is deleted between its click and the second rt-zone click.
Hmm...

----
Tue Dec 18 16:45:06 GMT 2007


 fill_close_option_menu_with_all_molecule_options() does not use
 GINT_TO_POINTER().  It should.


----
Tue Dec 18 16:46:52 GMT 2007

  separate out the widget parts of c-interface.h into
  c-interface-gui.hh (or should it be .h? - Yes I think it should be
  because those functions will also be called by callbacks.c)

-----
Thu Dec 20 20:25:45 GMT 2007

    JED: What I want is function to remove alt confs of atoms in a
    residue which only have one atom of that type in the residue.
    And set the occupancy to 1 at the same time.

    Give me a key binding to run this function on the active
    residue. And re-display the residue-info dialog.

* Done: (sanitise-alt-confs)

---
Fri Dec 21 13:51:31 GMT 200

 c-interface.cc:
 * Copyright 2004, 2005, 2006, 2007 by Bernhard Lohkamp
coot_utils.py:
# Copyright 2004, 2005, 2006, 2007 by Bernhard Lohkamp

 Remove that - unless there is good reason not to.

-----
Fri Dec 21 14:08:25 GMT 2007

   Bill report of starting up with GUI, but no scripting window available:




Latest coot, I am getting this when I try to open the python scripting
window (used to work):

Traceback (most recent call last):
 File "<string>", line 1, in <module>
NameError: name 'coot_gui' is not defined


I think everyting loaded:

DEBUG:: stating pydirectory /sw/share/coot/python
INFO:: loading coot.py from /sw/share/coot/python/coot.py
Running python script /sw/share/coot/python/coot.py
INFO:: coot.py loaded
INFO loading coot python modules
Running python script /sw/share/coot/python/coot_load_modules.py
Loading: coot_utils.py
Loading: filter.py
Loading: coot_lsq.py
Loading: shelx.py
Loading: get_ebi.py
Loading: local_code.py
Loading: hello.py
Good Evening Wgscott, Welcome to Coot.
(set-display-intro-string "Good Evening Wgscott, Welcome to Coot.")
Loading: mutate.py
Loading: refmac.py
Loading: libcheck.py
Loading: gap.py
Loading: fitting.py
Loading: raster3d.py
Loading: povray.py
Loading: generic_objects.py
Loading: ncs.py
Loading: cns2coot.py
Loading: tips.py
Loading: americanisms.py
Loading: group_settings.py 


Note that it is not starting up in python_gui mode.

Note also that pygtk is not needed for the python gui.

   Priority: Important.
   Bernhard to fix.
   (It doesn't happen to me, I get good results on searching for pygtk in configure).

   Try compile without pygtk.
   We should get python scripting window without pygtk.

* Discussed this with Bernhard.  He will try to untangle it.

-----
Fri Dec 21 15:11:40 GMT 2007

A valgrind thing to think about:

==15565== 496 (160 direct, 336 indirect) bytes in 1 blocks are definitely lost in loss record 264 of 402
==15565==    at 0x4C220C5: operator new[](unsigned long) (vg_replace_malloc.c:199)
==15565==    by 0x65F7293: Bond_lines_container::make_graphical_bonds() const (Bond_lines.cc:1642)
==15565==    by 0x4AD80C: graphics_info_t::flash_selection(int, int, std::string, int, std::string, std::string, std::string) (graphics-info-modelling.cc:757)
==15565==    by 0x4B3CE2: graphics_info_t::refine_residue_range(int, std::string const&, std::string const&, int, std::string const&, int, std::string const&, std::string const&, short) (graphics-info-modelling.cc:918)
==15565==    by 0x4B44DD: graphics_info_t::refine(int, short, int, int) (graphics-info-modelling.cc:842)
==15565==    by 0x53122B: refine_zone_with_full_residue_spec (c-interface-build.cc:5165)
==15565==    by 0x531308: refine_zone (c-interface-build.cc:5143)
==15565==    by 0x83E0EBB: _wrap_refine_zone(scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*) (coot_wrap_guile.cc:12815)
==15565==    by 0x8FF5107: scm_gsubr_apply (in /home/paule/x64/lib/libguile.so.17.1.1)
==15565==    by 0x8FDE10C: scm_dapply (in /home/paule/x64/lib/libguile.so.17.1.1)
==15565==    by 0x8FE2801: deval (in /home/paule/x64/lib/libguile.so.17.1.1)
==15565==    by 0x8FE2F44: deval (in /home/paule/x64/lib/libguile.so.17.1.1)
==15565== 

-----
Mon Dec 24 16:48:47 GMT 2007
	
  BL has introduced split in on_open_coordinates1_activate.  Needs
  rationalizing.

-----
Sun Dec 30 10:29:43 GMT 2007

   Google for this from the office:

   The Lancet : Ten pieces of free software every doctor should have

-----
Mon Dec 31 16:44:49 GMT 2007

   on doing a refinement: 

INFO:: graphics_info_t::clear_up_moving_atoms...
clearing intermediate object...
c-interface clear_up_moving_atoms...
INFO:: graphics_info_t::clear_up_moving_atoms...
WARNING:: attempting to delete NULL moving_atoms_asc.atom_selection
Ignoring. 
attempting to delete NULL moving_atoms_asc.mol
ignoring 
clearing intermediate object...

    (ugliness)

-----
Mon Dec 31 16:53:00 GMT 2007

   For 0.5:

   o Tests for chi angles: 
       Make some residue that you know the chi angles of.
          Generate values, compare.
       Chi angles of a Water
       Chi angles of 3GP

     Where should this code go?  In a new testing.cc/testing.hh?  
     test_chi_angles(imol_3gp, imol_ref, imol_water);
        and the actual target values are embedded in the test code.
        Hmm...  Well, I could try it.

     Note to self:  Write the test first!

* Done.  In test-ligand

----
Fri Dec 28 17:12:58 GMT 2007

    how do I test torsion general?

    I can use test-ligand, for sure.

    maybe, I'll libify the test, so coot main can use it.
    Hmm...

    So we make molecules every n*10 degress
 
    rotate the bond the other -n*10.

    Overlap results.  
       Test for closeness.

-----
Mon Dec 31 19:48:09 GMT 2007

   interesting?
   FBSS:
   http://cisrg.shef.ac.uk/projects/fbss//man_contents.html

-----
Mon Dec 31 23:08:32 GMT 2007

  molecule_info_class_other.cc has:
  do_accept_reject_dialog("Alt Conf Split", coot::refinement_results_t());
  Yeuuuuucch!

  I see why I did it I think - it was cheap :)
  It is in split_residue_internal(), so the resulting value from this
  function needs to be propogated back up through split_residue()

-----
Wed Jan  2 18:57:42 GMT 2008

   Funny shapped protein?

	1l7l 

-----
Mon Jan  7 08:48:31 GMT 2008

   Export chi angles for a residue spec to scripting layer.
   Can test from there.

---
Tue Jan  8 04:12:57 GMT 2008

  Make a test for alt conf split a residue?
  What is the test?  There is not functions at the scripting layer:
  add_alt_conf( [residue spec] ) 
  or 
  split_residue ( [residue spec] ) 
  
  Hmmm...

-----
Tue Jan  8 07:57:19 GMT 2008

    Transfer transpose matrix to Coot from Chart.
    Use it, like Bernhard does on the nudge view buttons.

* Done

----
Tue Jan  8 09:40:44 GMT 2008

   A graph tool to represent NCS differences

* Done.

----
Tue Jan  8 09:41:15 GMT 2008

   Rewrite sequence view (from scratch)

----
Tue Jan  8 10:42:12 GMT 2008

this needs filling
set_monomer_restraints(). 
and the function should be in c-interface-build.cc I think.


---
Tue Jan  8 11:56:22 GMT 2008

   A script that identifies badly fitting residues, 
   (by density fit and geometry and rotamer say - these residue
   attributes need to be exported to scripting surface)

   make a list of residues
     view to residue
     rotate view
     try rotamer search
     does it fit now?
     if not
        delete residue
        add-terminal-residue
        mutate to previous residue type
     rotate view

   Do it in scripting

----

    PSK18 has tunnel for C16, long chain fatty acid 
    Ruckmini 2004
    Might be fun to do an annimatation of this.

    Priority: (Fun), Low
 

----
Tue Jan  8 13:06:56 GMT 2008

    Testing fixed atoms:

    need to:
    create a bad rotamer
    refine residue
      {generates intermediate atoms}
    fix a wrongly positioned atom
    drag another atom
      {fixed atom doesn't move}
      {continue refinement}
    accept refinement
    get atom attributes for fixed atom
    check that it is where it should be (fixed position)
   

----
Tue Jan  8 13:14:48 GMT 2008

    Rotamer funtion:
    set_rotamer_at_residue(imol, chain_id, resno, ins_code, alt_conf, rotamer_name);
    where rotamer name is "t" or "tpp" etc.

-----
Wed Jan  9 00:30:49 GMT 2008

    For 0.5:

    Do a validation on a fitted 3GP.  Is the most deviant geometry the
    one shown in the tooltip?

    Priority: High.

---
Wed Jan  9 07:03:22 GMT 2008


  Look up the "Ramachandran" plot for torsions around the glycosidic
  bond in carbohydrate linkages

  phi vs psi

  (-90, 60) is the big peak

  around c2-o and o-xxx on the other subunit

  Look For The Database analysis of Glycosylation

  1f31 has axial carbohydrate in ganglioside, a mistake maybe?

-----

molecule_class_info_ncs: has this construction:

   for(unsigned int ighost=0; ighost<ncs_ghosts.size(); ighost++) {

      if (ncs_ghosts_have_rtops_flag == 0) {
	 // std::cout << "   %%%%%%%%% calling fill_ghost_info from "
	 // std::cout << make_dynamically_transformed_maps "
	 // << std::endl;
	 fill_ghost_info(1, homology_lev); // fill the rtops and set the flag
      }

is that right?

-----
Thu Jan 10 08:19:19 GMT 2008
	
   write a testing.cc test for 
	std::vector<std::pair<int, int> >
	coot::util::pair_residue_atoms(CResidue *a_residue_p,
				       CResidue *b_residue_p) {

    passing different residue types

----
Thu Jan 10 10:43:33 GMT 2008

  Aspirin and paracetamol are simple.

  See if coot can fit them (do they exist in the EDS?).  Add to test
  suite?

----
Fri Jan 11 08:40:05 GMT 2008

   Look at fitting strands at low resolution.
   an OMP pore protein perhaps.

   One for the test suite?

-----
Fri Jan 11 08:51:06 GMT 2008

   Make the default bond thickness thicker!

* Done.

----
Fri Jan 11 10:40:28 GMT 2008

   Transfer map functions to molecule-class-info-maps.cc

* Done.

-----
Sun Jan 13 14:50:19 GMT 2008

    A function to return the density fit score of a residue.  Can be
    done in scripting of course, now (not sure if that is for the
    best)

---
Sun Jan 13 14:51:14 GMT 2008

   In ligand fitting, I want to see the scores and the positions of
   the best fits in the other orientations.  The B factors should be
   set to the ligand fit score for this testing.  Perhaps the ligand
   fitting should return a vector of std::pair<name, mol>s.  Best fit,
   ori1, ori2, etc.

   So maybe set a developer option to do this.

---
Sun Jan 13 15:11:30 GMT 2008

   Residue info should also contain phi psi values and a percentile
   (like molprobity)

   How about a graph of Ramachandran percentile? 


----
Sun Jan 13 15:36:10 GMT 2008

    Test that skip-to-next-ncs-chain works with a2b2 hemoglobin.  


----
Sun Jan 13 18:42:32 GMT 2008

    How To Add Another Geometry Graph
  
    in geometry_graphs.hh:
    --------------------------
    Add new type to geometry_graph_type 

    in geometry_graphs.cc
    --------------------------
    In the coot::geometry_graphs::geometry_graphs(...) constructor, add title

    in graphics-info-graphs.cc
    --------------------------
    graphics_info_t::update_geometry_graphs(...) 
    Add a function to update new graph.

      coot::geometry_graphs *gr =
	 geometry_graph_dialog_to_object(residue_density_fit_graph[imol_moving_atoms]);
	 std::vector<coot::geometry_graph_block_info_generic> dv =
	    geomtry_graph_xxx_from_mol(moving_atoms_asc_local,
				 imol_moving_atoms,
				 Imol_Refinement_Map());
	 gr->update_residue_blocks(dv);

    Create a function:
    void graphics_info_t::xxx_graphs(int imol)...


    in c-interface-xxx.cc
    ----------------------
      void validation_graph_XXX_mol_selector_activate (GtkMenuItem     *menuitem,
						      gpointer         user_data);
   in main.cc
   ----------
   Add create_initial_validation_graph_submenu_generic
   The menu name (the first string) must match the names generated by the glade

   in callbacks.c
   --------------
  char *type = "density-fit";
  GtkWidget *menu = lookup_widget(GTK_WIDGET(menuitem), "density_fit_analysis1");
  if (menu) { 
    add_on_validation_graph_mol_options(menu, type);
  } else { 
    printf("failed to get menu in on_density_fit1_activate\n");
  }
    
   in c-interface-validate.cc
   -------------------------

   Add another option to add_on_validation_graph_mol_options()

   The validation_type must match that from callbacks.c
   The sub_menu_name must match that from main.cc 
     create_initial_validation_graph_submenu_generic()

   Also unset_geometry_graph(GtkWidget *dialog) needs to know about the new graph

   in graphics_info_graphs.cc 
   --------------------------
   set_validation_graph need to know about the new graph


-----
Mon Jan 14 13:22:33 GMT 2008

    testdbmain is broken.


----
Mon Jan 14 13:22:48 GMT 2008

    Really test out Ca Zone -> Mainchain.  

    It seems to me that most of the time it works well, but there are
    regions where it does not. e.g. CA - N bonds of less than 0.5A.
    I am thinking that that can only be coming from a bug.

    Think about how to test that. 

    Very few reference structures?

    We need to know how many fragment are contributing to each
    position.


-----

   configure should check for scsh.
   only if it has it should it be used for the documentation.

----
Thu Jan 17 09:50:46 GMT 2008

    think about user-defined bond, as seen on Phenix list recently:
    e.g.:

    add-user-defined-bond(int imol, atom_spec atom1, atom_spec_atom2, float dist_ideal, float sigma);

----
Thu Jan 17 10:14:18 GMT 2008

    Tinker with a libglade project.

-----
Thu Jan 17 11:56:57 GMT 2008

    ubuild, ie. without python out the box please.  
    i.e. coot_wrap_python.cc needs to be not compiled somehow.

   Priority: Important.


----
Sat Jan 19 11:35:01 GMT 2008

    Build GTk2 on YSBL.

chihiro fails: no GL include files.
dewinter fails: in mmdb: C++ compiler can't compile

bubbles fails: gtk2 version is 2.0.6, needs 2.2.0
bragg3 fails: can't compile a working guile 1.8.3

----
Sat Jan 19 11:56:58 GMT 2008

    Molprobity interaction dots around a ligand.  
    Surely that would be super cool?

    Yes it would.

-----
Sat Jan 19 16:09:27 GMT 2008

    Read charges from dict and calculate and display diapole moments

----
Mon Jan 21 13:04:37 GMT 2008

    Now I am compiling gtk2+ becuase the system gtk2+ on cycle (FC4)
    is too old to complile Coot.

    I am doing it on cycle in ~/compile
    I created setup pango-glib-gtk2 with PKG_CONFIG settings.
    ./configure --prefix=$HOME/autobuild/Linux/Coot-0.4

    glib 2.12.13 straightforward
    pango 1.16.4 straightforward
    fine.  
    Oh for gtk2 though, cairo is not an option.  It is a must.  Baah.
    That was not clear in www.gtk.org.

    So get cairo 1.4.14 from http://cairographics.org/releases/

    which means get glitz from http://cairographics.org/snapshots/
    (gotta have glitz, right?)

    glitz-0.5.6 
    ./configure --prefix=$HOME/autobuild/Linux/Coot-0.4
    Compiles OK.

    back to cairo
    ./configure --prefix=$HOME/autobuild/Linux/Coot-0.4 --enable-glitz
    This displays a warning about using glitz, but configures OK.
    Compiles OK.

    back to pango:
    now with cairo in place, pango's configure finds cairo
    ./configure --prefix=$HOME/autobuild/Linux/Coot-0.4
    Compiles OK.

    back to Gtk2+:
    glib version test failed, so need to add LD_LIBRARY_PATH too (to the setup file)
    ./configure --prefix=$HOME/autobuild/Linux/Coot-0.4
    Compiles OK.

    Fine.  But Coot configure fails on gtlglExt

    http://downloads.sourceforge.net/gtkglext/gtkglext-1.2.0.tar.gz?modtime=1139081517&big_mirror=1
    back to GtkGLExt:
    ./configure --prefix=$HOME/autobuild/Linux/Coot-0.4
    Compiles OK.

    Coot now compiles OK.  And runs! Great.
    But - it doesn't load svgs.

    OK, back-up. 

    config.log for cairo tells us we need a newer librsvg:
    http://ftp.gnome.org/pub/GNOME/sources/librsvg/2.20/
    http://ftp.gnome.org/pub/GNOME/sources/librsvg/2.20/librsvg-2.20.0.tar.gz (2 days old at 
    time of writting)

    ./configure --prefix=$HOME/autobuild/Linux/Coot-0.4
    Warns us that we are not compiling with libcroco.
    Oh well.

    OK, now librsvg is in place configure finds it.
    Rebuilding cairo
    Install.

    Fine.  Now coot loads its svgs.  Yay!

    PyGtk needs pycairo


    Correct build order:
    glitz
    glib
    librsvg
    cairo
    pango
    gtk+-2
    glkglext

Hmm...


checking for LIBRSVG... configure: error: Package requirements (        gdk-pixbuf-2.0 >= 1.3.7         glib-2.0 >= 2.12.0      libxml-2.
0 >= 2.4.7      pangoft2 >= 1.2.0       pangocairo >= 1.2.0     cairo >= 1.2.0  cairo-png >= 1.2.0      fontconfig >= 1.0.1
        freetype2) were not met:

Requested 'pangoft2 >= 1.2.0' but version of Pango FT2 is 1.1.1

http://www.linuxfromscratch.org/blfs/view/svn/x/cairo.html

  Note that the GTK, Poppler and librsvg packages are circular in that
  using them for test suite coverage requires installing cairo first,
  then installing the desired package(s), then installing cairo again.

Now on to guile-gtk:

A new guile: 1.8.3:

./configure --prefix=/lmb/wear/emsley/guile-1.8.1 CFLAGS="-I/lmb/wear/emsley/guile-1.8.1/include -O" LDFLAGS=-L/lmb/wear/emsley/guile-1.8.1/lib

if I use -g instead of -O, I get:

$ guile
Backtrace:
In unknown file:
     ...
   ?: 123  [list apply ...
   ?: 124*  [list lambda (syntmp-dummy-3 syntmp-name-4 syntmp-rules-5) ...
   ?: 125*   [syntmp-chi-151 #(syntax-object # #) (# # # # ...) ...
   ?: 126*    [syntmp-make-binding-wrap-132 (# # #) ("i" "i" "i") (())]
   ?: 127     (if (null? syntmp-ids-949) syntmp-w-951 ...)
   ?: 128     [cons () ...
   ?: 129*     [cons ...
   ?: 130*      (let* ((syntmp-labelvec-952 #)) (let* (#) (let # #)))
   ?: 131       (let* ((syntmp-n-953 #)) (let (# #) (begin # #)))
          ...
   ?: 132       (begin # #)
   ?: 133*      [syntmp-f-956 (# # #) 0]
   ?: 134       (if (not #) (call-with-values # #))
          ...
   ?: 135       [call-with-values #<procedure #f ()> #<procedure #f #>]
   ?: 136       (@call-with-values (producer consumer))
   ?: 137*      [#<procedure #f ()>]
   ?: 138*      [syntmp-id-sym-name&marks-116 #(syntax-object dummy #) (())]
   ?: 139       (if (syntmp-syntax-object?-101 syntmp-x-980) (values # #) ...)
   ?: 140*      [syntmp-syntax-object?-101 #(syntax-object dummy (# # # shift ...))]
   ?: 141       (and (vector? syntmp-x-1008) (= (vector-length syntmp-x-1008) 3) ...)
   ?: 142*      (vector? syntmp-x-1008)


guile-gtk-2.1:

./configure --prefix=$HOME/autobuild/Linux/Coot-0.4 --without-gtkgl

-----
Mon Jan 21 15:49:32 GMT 2008

    reference for EDS:

 Title: The Uppsala Electron-Density Server
Author(s): Kleywegt GJ, Harris MR, Zou JY, Taylor TC, Wahlby A, Jones TA
Source: ACTA CRYSTALLOGRAPHICA SECTION D-BIOLOGICAL CRYSTALLOGRAPHY 60: 2240-2249 Part 12 Sp. Iss. 1, DEC 2004 
  
* This is in the manual already.


----
Mon Jan 21 17:07:30 GMT 2008

    Refinement (with chiral volumes) on a residue that does not have
    the atoms for a chiral centre fails with E_NO_PROGRESS.

    Priority: Critical


----
Sat Jan 26 14:14:39 GMT 2008

    Don't forget to distribute val.pdb.  Else the tests will fail.

----
Sat Jan 26 15:49:27 GMT 2008

   o Copy fragment (new-molecule-by-atom-selection) does not
     calculate the NCS.  Maybe it should.


-----
Sat Jan 26 20:49:44 GMT 2008

    for scroll wheel map template see cinterface.cc:1003

    set the scroll wheel map correctly from a cns map.
    molecule-class-info-other.cc:6288

    set the scroll wheel map correctly from a phs map.
    molecule-class-info.cc:3203

    set the scroll wheel map correctly from a sfcalc map
    molecule-class-info.cc:3538

    set the scroll wheel map correctly from a cif nfofc map
    molecule-class-info.cc:3896

    set the scroll wheel map correctly from a cif phs
    molecule-class-info.cc:4018

    add tests for reading phs and cif files, 
    test that they return sensible molecule numbers.

    add a test for creating a new molecule from water fitting.
    using void find_waters(imol_for_map,  imol_for_protein,  new_waters_mol_flag, 
		 sigma_cut_off, show_blobs_dialog);


    add a test that, when a map is made with refmac params, the refmac
    params are correct.  May need an accessor function (return a SCM)

    add a test for both a map and a molecule that the new molecule
    number is correct (should match graphics-n-molecules, I think).

    test try setting an atom attribute of a bugus (-1, 888) molecule number


    think about what to do with:
    graphics_info_t::expand_molecule_space()

    why in this function:
    handle_cns_data_file_with_cell(const char *filename, int imol..)
    is imol  passed?

------
Tue Jan 29 14:49:39 GMT 2008

    Expose overlap ligands to the gui.

-----
Wed Jan 30 13:57:25 GMT 2008

   Randy Read: 

   "I've just run across what seems like a bug for proteins with
   hydrogens, in coot 0.4.1.  I've just been refitting something
   refined with all hydrogens on.  If I choose a new rotamer for some
   of them (e.g. Lys), it seems everything is fine.  However, if I
   choose a new rotamer for Arg, Asn, Val or Leu, it looks fine until
   I click "Accept", and then (at least some of) the hydrogens get
   left behind in their old positions"

   Hmmm... Hydrogen name problem, I'd guess

   Need to test on a protein that comes from molprobity.

* Randy Agrees.

-----

   pygtk needs pycairo:

   http://cairographics.org/releases/pycairo-1.4.12.tar.gz

   If you build your own pkg-config, you need to add
   /usr/lib/pkgconfig to PKG_CONFIG_PATH environment variable.


----
Wed 30 Jan 2008 14:12:55 GMT


on compiling freeglut on impala:


 -o CallbackMaker-CallbackMaker.o `test -f CallbackMaker.c || echo './'`CallbackMaker.c
/bin/sh ../../../libtool --mode=link gcc  -I/usr/X11R6/include   -o CallbackMaker -export-dynamic ../../../src/
libglut.la CallbackMaker-CallbackMaker.o  
mkdir .libs
gcc -I/usr/X11R6/include -o .libs/CallbackMaker CallbackMaker-CallbackMaker.o  -L../../../src/.libs -lglut -lm 
-L/usr/X11/lib -lGL -lGLU -lXxf86vm -lXext -lX11 -lXdmcp -lXau
ld: cycle in dylib re-exports with /usr/X11/lib/libGL.

http://www.mail-archive.com/fink-devel@lists.sourceforge.net/msg16105.html
  
   Problem solved - I have it working.
   To the ConfigureParams section of the info file I added:

   LDFLAGS="-dylib_file /System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib:/System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib"

PE: This worked for me too:


./configure --prefix=/Users/paule/autobuild/Darwin/coot-pre-release-gtk2 --disable-warnings LDFLAGS=-dylib_file /System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib:/System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib

for OpenGL (in gtkglext building) 

 ./configure --prefix=/Users/paule/autobuild/Darwin/coot-pre-release-gtk2 --disable-warnings LDFLAGS="-dylib_file /System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib:/System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib"

------
Wed 30 Jan 2008 14:50:08 GMT

   cairo depends on freetype
   
   (but it fails during compile, not cofigure, baah)

   http://download.savannah.gnu.org/releases/freetype/

* Done.  in build-it-gtk2 now.


------
Thu Jan 31 00:48:53 GMT 2008

   Alice Dawson:

   1:
   > I can't set the occupancy to a number greater than 1 in the
   > residue info box.  This is quite annoying when fiddling with
   > disordered residues - e.g. when I delete CA, B for a residue the
   > remaining CA still has occupancy -21 which I can't alter in
   > Coot..  Also when I add any molecules (waters, ligands) they are
   > added with an occupancy of 1 not 11.

   2:
   > An item for a wishlist: my protein has 1000 residues in the
   > asymmetric unit and it's very difficult to keep track of the free
   > variables.  Including the FVAR number and value in the Residues
   > with Alt confs box would be really helpful.

   1: Priority High

   2: Priority Medium

----
Thu Jan 31 00:49:48 GMT 2008

   Francis Reyes:
 
   > I'm trying to merge / append two molecules. However, when I click
   > on Merge, I get ...  INFO:: adding molecule 6 to merging list

   > ... and that's it..

   > Yet the residue does not appear as a new chain in the molecule
   > into which I'm merging?  Earlier versions of coot actually show
   > INFO:'s telling you which chain it made, etc.

----
Thu Jan 31 07:49:51 GMT 2008

   command line tool to get OS version on Mac?

   These of use?

/bin/arch              = unknown
/usr/bin/arch -k       = unknown
/usr/convex/getsysinfo = unknown
/usr/bin/hostinfo      = unknown
/bin/machine           = unknown
/usr/bin/oslevel       = unknown
/bin/universe          = unknown

------
Thu Jan 31 08:32:26 GMT 2008

    Select rotamer button box has an OK button that needs to be homogenized.

    Priority: High

----
Thu Jan 31 20:12:56 GMT 2008

   Clemens (on density fit graphs):

   Since I find it easier to ask others for those mental gymnastics
   (aka Paul in this case) than doing it myself: how about having all
   those graphs always starting at number 1 - or at the lowest residue
   number of any chain?

   Yes the lowest number in any chain.  That is, if the difference is
   not too large (like 100 or so)

   Priority: Medium.  Do before 0.6.

-----
Fri Feb  1 19:03:34 GMT 2008

    Need a test for merging molecules.
    merge-molecules is used in the rna test.
    But let's make more of a torture test and test the resulting molecule

------

	look in ejd/2a4d something and residue 
	218A rotamers, fails.
	But works after regularization

-----
Wed Feb  6 14:05:16 GMT 2008

    cc-interface-build.cc has the following bit of code:

	 if (p.size() > 0) {
	    // install new molecule(s) that has been rotated.
	 }

   Should fix/fill.



-----
Wed Feb  6 14:58:14 GMT 2008

   Dave Schuller:

   In Coot 0.4.1, I have a molecule displayed as "Ca + ligands." After
   I ran some regularisation, the color of my molecule changed! I
   switched the molecule display mode to bonds, then back to Ca +
   ligands and everything was OK again. Interesting "feature."

   Priority: High:
   Difficulty: Easy.

-----
Wed Feb  6 16:15:31 GMT 2008

    On mutate and insert residue we can additionally set the seg-id of
    the new atoms (given that the current/previous residue is sensible).

    Priority: Medium:
    Difficulty: Medium/Easy


----
Wed Feb  6 16:17:51 GMT 2008

    Doug Kuntz:

    Make a method to save the symmetry colour.

    Priority: High.  Do it for 0.5
    Difficulty: Easy 


---
Thu Feb  7 12:24:59 GMT 2008

Doug Kuntz:

> > > 2) The selection of active molecules seems screwed up.
> > >
> > > I can now only have one active molecule.  Even though all the active 
> > > boxes are nice and pink in the Display Manager window.  To move a 
> > > molecule I have to ensure all the others are turned to inactive. But 
> > > there seems no way to make more than one molecule active at a time. If 
> > > I want to compare two superimposed molecules for distances between 
> > > them  at a particular atom I can no longer  pick both atoms

      Ingo Korndoefer reported something similar.


-----
Wed Feb  6 23:18:23 GMT 2008

    After graphs moved over to molecules class, 

    Need to test Ramachandran plot and sequence view for molecules


-----
Thu Feb  7 00:00:14 GMT 2008

    Why is there a delete plot in 
    void
    graphics_info_t::set_dynarama_is_displayed(GtkWidget *dyna_toplev, int imol)

---
Fri Feb  8 20:37:43 GMT 2008

    If you are going to edit restraints, wouldn't it be a good idea to
    write a restraints file (mmdb write mmCIF interface).


---
Sat Feb  9 22:40:38 GMT 2008


A FC3 machine:


bunyip ~> pkg-config --modversion gtk+-2.0
2.4.14

A FC4 machine:

cycle ~> pkg-config --modversion gtk+-2.0
2.6.7 

-----
Mon Feb 11 08:26:02 GMT 2008

GDK_PIXBUF_MODULE_FILE=$HOME/autobuild/Linux-bragg1.chem.york.ac.uk-pre-release-gtk2/etc/gtk-2.0/gdk-pixbuf.loaders

export GDK_PIXBUF_MODULE_FILE

----
Mon Feb 11 13:02:51 GMT 2008

    There was a problem of left-over muddled gtk libs and dependencies
    in coot-Linux-fedora-5-gtk.

    Which caused problems in startup.  If I ran the slim function by
    hand to a clean directory, it worked (woop!)

    diagnosed by using 
    diff -r dir-1 dir-2

    Now rebuilding fc5 (bragg3) from scratch.

    Need to test that there are no dependencies on
    autobuild/Linux-bragg3-pre-release-gtk2 when its done.

    Question:  are pygobject and pycairo needed for pygtk 2.6.3?


-----
Tue Feb 12 01:37:53 GMT 2008

    coot-real from kalypso ~/Projects/build-coot-fc4-gtk2/src> ./coot-real

    Try to open an mtz file from tutorials, 

    on changing to  "tutorials" in the ccp4i project option menu:

** (lt-coot-real:32355): WARNING **: Widget not found: map_name_fileselection1
WARNING:: failed to find filesection in option_menu_refmac_ccp4i_project_signal_func



------
Wed Feb 13 10:48:53 GMT 2008

   The flatland thing on NCS happened without drawing NCS ghosts first.

   So, is an NCS matrix being applied when it shouldn't? NCS ghost are
   marked as filled, but they are not?

-----
Thu Feb 14 20:55:59 GMT 2008

   handle_symmetry_colour_change(int mol, gdouble* col)
 
   is the same functionality as 

   set_symmetry_colour().

   Rationalize it.

   Also get_map_colour and get_map_color_gdkcolor needs rationalizing.


-----
Thu Feb 14 21:02:39 GMT 2008

    I am on a symmetry related atom. I want a "go to atom (relative
    main molecule atom)"


  g_signal_connect ((gpointer) map_colorbutton, "color_set",
                    G_CALLBACK (on_map_colorbutton_color_set),
                    NULL);

-----
Fri Feb 15 01:08:40 GMT 2008

    When a GtkColorButton is clicked, we get a GktColorSelectionDialog
    (or some such).  How do I control the colour selection dialog so
    that it work in continuous update mode, so that signals are sent
    continuously as the color selection change

----
Sun Feb 17 00:14:05 GMT 2008

    in simple-restraint init_from_mol() we do
    SelHnd_atom = mol->NewSelection();

    do we ever give SelHnd_atom back?


-----
Mon Feb 18 12:36:34 GMT 2008

    delete-residue-range seems to do 2 backups.


-----
Mon Feb 18 12:38:02 GMT 2008

DEBUG::  display toggle of molecule: 0
DEBUG::  display toggle of molecule: 2


-----
Wed Feb 20 04:36:04 GMT 2008

    Read in 1qex from EDS.

    calculate ghosts

    density fit.

    rescale 0.5

    recale 0.3

    click on first residue in water chain

       -> crash

INFO:: canvas size based on 3 chains with max length 302
INFO:: canvas size: 3220 270
DEBUG:: min_resno: 1 max_resno: 288 chain_number 0
DEBUG:: min_resno: 1 max_resno: 288 chain_number 1
DEBUG:: min_resno: 1 max_resno: 302 chain_number 2
INFO:: trying to centre on:  1  O  
(set-go-to-atom-molecule 0)
Segmentation fault (core dumped)


----
Wed Feb 20 05:27:36 GMT 2008

    302 EPE in 1QEX.  Look at the bonding to the Sulfur.  One of the
    oxygens has a cross. Fix?

----
Wed Feb 20 11:27:37 GMT 2008

   Here: 
   void
   graphics_info_t::start_baton_here()

   we do: 
      int *imol_copy = new int;
      *imol_copy = imol_for_skel;
      gtk_object_set_user_data(GTK_OBJECT(w), (char *)imol_copy);

   should use GINT_TO_POINTER

-----
Thu Feb 21 03:52:58 GMT 2008

    save symmetry coordinates and save coordinates should have an
    scripting history expression.

-----
Thu Feb 21 06:52:25 GMT 2008

    display manager should resize better.
    Autosize up to a certain size (ysize).

----
Thu Feb 21 06:53:17 GMT 2008

    Frank says make a selection in the go to atom window and use it to
    make new molecules.  Or cut/paste.

    Idea needs more details.

----

ARP/wARP peptide torsion from pseudo bonds

2 torsions from CA-O-CA-O-CA  -> "American View of the World"


------

   when the fcf file is in someone else's directory, the writing of
   the cif file fails.

-----

 D. Gayathri 
gaya_raju@rediffmail.com, dcbgaya@gmail.com

 
------

   Crash by Martyn

   do unmodeled blobs.
   dont close results
   
   get unmodelled blobs dialog again
   and martyn said that he didn't click...

-----
Fri Feb 22 08:36:20 GMT 2008

    Note to self:

    Redo the ligand fitting slides.

    Easy to do 
   
----

 Sergs' presentation:
 Autosolve ligand detection Mooji (?) 2006?


 Use cluster volume vs contour threshhold plot (of the difference map)
 to detect the cluster (volume decreases with contour level less for
 real ligand than noise.

 and use the blob that is closest to the protein (Frank is sceptical
 about that).
 
 there is some sort of shape function too (I missed how).  ah, 7
 statistics, sufrace to volume ratio, eigen values etc.


------
Fri Feb 22 09:31:35 GMT 2008

    Import an ATP.

    Regularize it.

    furanose ring is flattened... why?

    Bug, I'd say

    On testing, maybe not.  Is the restraints.
    Need to get other restraints for ATP.


----
Fri Feb 22 09:45:25 GMT 2008

  Get a list of ATPs from pdb to fit *this* particular ATP

  (will do conformation search for us).

  Report the structure that matched.

  In fact, download the matchng structure and overlay it on the
  ligand.


-----

	Can I get to MSDchem via SOAP or some such?
	Don't bother - use sbase.


----

	Add a water over a symmetry atom.

	go to real space of symmetry atom.

	move the real space atoms a bit.

	shift left on sym water. Fails to label.



-----
Sun Feb 24 09:55:35 GMT 2008

    Try running baton mode without first making a skeleton.
    
    It doesn't work properly.

    Priority: Medium
    Difficulty: Medium


-----
Fri Feb 29 16:33:00 GMT 2008

    more on compiling on impala


If I copy the command line from configure for pygtk and pyobject

[this is from pygtk]
./configure --prefix=/Users/paule/autobuild/Darwin/coot-pre-release-gtk2 --with-pango --with-pangocairo PYGOBJECT_CFLAGS=-I/Users/paule/autobuild/Darwin/coot-pre-release-gtk2/include/pygtk-2.0 PYGOBJECT_LIBS=-L/Users/paule/autobuild/Darwin/coot-pre-release-gtk2/lib

then it configures on the command line.  But not from script (glib test
program problems).  Grrr.

----
Wed Feb 27 09:18:50 GMT 2008

    Residue Info on a residue with alt confs.  
    Problem in CHI angles?
 
-----
Wed Feb 27 15:38:28 GMT 2008

    At the end of build-it-gtk2-simple:

    mkdir: cannot create directory
    `/home/paule/autobuild/coot-Linux-i686-ubuntu-7.10-gtk2/include/python2.5':
    No such file or directory

   indeed the include dir is not there.


----
Sat Mar  1 06:11:26 GMT 2008

    Parse USER MODs
    somewhere I have some code, I started on this?

    on penelope, I guess.


----
Wed Mar  5 17:52:08 GMT 2008

  pick_info
  graphics_info_t::find_atom_index_from_goto_info(int imol) 

  very convoluted!?

  Why not use graphics_info_t::molecules[imol].atom_to_atom_index()? 

----
Thu Mar  6 13:31:13 GMT 2008

    Get rid of (MyCMMDBManager *) 

----
Fri Mar  7 13:40:51 GMT 2008

    Test for guile-gtk is failing (i.e. guile-gtk2.1 is being built)
    on bragg3.

----
Fri Mar  7 14:48:47 GMT 2008

    Useful fun.  
    Write a parse for gdk/gdkkeysyms.h
    if field-1 is #define if NF == 3
    split field-2 after GDK_ and decode field-3 to decimal
    
    now we can have a function gdk-key-symbol->key

    e.g
    (gdk-key-symbol->number "Super_R")
    -> 2354 [or some such]

    now change graphics-general-key-press-hook to test for a string,
    and if it is, run gdk-key-symbol->number.

----
Wed Mar 12 12:00:50 GMT 2008

    revisit --as-needed, Donnie B.

----
Thu Mar 13 14:31:37 GMT 2008

   consider install_name_tool when installing mac binary?

----
Thu Mar 13 16:15:10 GMT 2008

    Additional representations "button" should (also?) be on the
    Display Control.

----
Sat Mar 15 09:30:11 GMT 2008

    Herb Klei: after a superposition (LSQ say) the moving molecule
    should take on the cell/symmetry of the target/reference molecule.

    When you do LSQ superpositions in COOT, can you make it so that if
    one does not specify residue ranges, all residues in common are
    assumed.  It seems silly to need to type in 1 and 999 twice just
    to run.  Also, I would make Calpha the default (can we do so
    through .coot file?).

----
Fri Mar 14 00:49:26 GMT 2008

    poly-ala in mutate.scm is not documented.  why?


----
Fri Mar 21 10:54:01 GMT 2008

   INFO::    Raising window e.g for display control:

   if (widget) { 
      // raise/uniconify (or whatever) what we have:
      // 
      if (!GTK_WIDGET_MAPPED(widget))
	 gtk_widget_show(widget);
      else
	 gdk_window_raise(widget->window);


-----
Mon Mar 24 23:51:58 GMT 2008

    NOTE to SELF:

    1ao0: fitting ADP fits the ADPs upside-down consistently.
    Hmmm....


----
Tue Mar 25 14:39:35 GMT 2008

    clear this junk:

BEFORE 0x1eaea170
DEBUG:: start of delete-residue-with-altconf n-atoms: 1465
DEBUG:: End of delete-residue-with-altconf n-atoms: 1465
failed to delete residue atoms A 88 :A:
AFTER  0x1eaea170
PASS: Deleting (non-existing) Alt conf and Go To Atom [JED]

* Done.


----
Tue Mar 25 15:01:27 GMT 2008


==4411== Use of uninitialised value of size 8
==4411==    at 0x1C2132DE: clipper::Util::rad2d(double const&) (clipper_util.cpp:156)
==4411==    by 0x1C1FE9FC: clipper::Cell_descr::format() const (cell.cpp:81)
==4411==    by 0x1BD9F301: coot::ShelxIns::debug() const (coot-shelx-ins.cc:1233)
==4411==    by 0x831BEF4: molecule_class_info_t::write_shelx_ins_file(std::string const&) (molecule-class-info-other.cc:4428)
==4411==    by 0x820DE1F: write_shelx_ins_file (c-interface-build.cc:6015)
[snip]
==4411==
==4411== Use of uninitialised value of size 8
==4411==    at 0x3B1AE12: std::ostream::operator<<(double) (in /usr/lib/libstdc++.so.6.0.4)
==4411==    by 0x1C20BED5: clipper::String::String(double, int, int) (clipper_types.cpp:69)
==4411==    by 0x1C1FEA30: clipper::Cell_descr::format() const (cell.cpp:81)
==4411==    by 0x1BD9F301: coot::ShelxIns::debug() const (coot-shelx-ins.cc:1233)


==4411== Source and destination overlap in strcpy(0x1F7DBC40, 0x1F7DBC40)
==4411==    at 0x1B90C51E: strcpy (mac_replace_strmem.c:210)
==4411==    by 0x1C393168: CSSVertex::SetVertex(CMMDBManager*, int, int, int, char*, int, char*, int, char*) (ss_vxedge.cpp:425)
==4411==    by 0x1C396354: CSSGraph::RepairSS(CMMDBManager*) (ss_graph.cpp:483)
==4411==    by 0x1C396865: CSSGraph::MakeGraph(CMMDBManager*) (ss_graph.cpp:379)
==4411==    by 0x1C38F61F: CSSMAlign::GetSSGraph(CMMDBManager*, int, int&) (ssm_align.cpp:147)
==4411==    by 0x1C38F72B: CSSMAlign::Align(CMMDBManager*, CMMDBManager*, int, int, int, int) (ssm_align.cpp:188)
==4411==    by 0x819179B: graphics_info_t::superpose_with_atom_selection(atom_selection_container_t, atom_selection_container_t, int, std::string, std::string, short) (graphics-info-superpose.cc:207)
==4411==    by 0x82483F3: superpose_with_atom_selection (c-interface-superpose.cc:201)


==4411== Mismatched free() / delete / delete []
==4411==    at 0x1B90B8CF: operator delete(void*) (vg_replace_malloc.c:155)
==4411==    by 0x1C3A0A3C: CSuperpose::_superpose(CSSGraph*, CSSGraph*, int&) (ssm_superpose.cpp:2683)
==4411==    by 0x1C3A1194: CSuperpose::SuperposeCalphas(CSSGraph*, CSSGraph*, int*, int*, int, CMMDBManager*, CMMDBManager*, int, int) (ssm_superpose.cpp:2424)
==4411==    by 0x1C38F7C8: CSSMAlign::Align(CMMDBManager*, CMMDBManager*, int, int, int, int) (ssm_align.cpp:200)
==4411==    by 0x819179B: graphics_info_t::superpose_with_atom_selection(atom_selection_container_t, atom_selection_container_t, int, std::string, std::string, short) (graphics-info-superpose.cc:207)
==4411==    by 0x82483F3: superpose_with_atom_selection (c-interface-superpose.cc:201)
==4411==    by 0x1BE55150: _wrap_superpose_with_atom_selection(scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*) (coot_wrap_guile.cc:23440)
==4411==    by 0x1BF757A0: scm_gsubr_apply (in /extra/paule/autobuild/Linux-kalypso-fc4-pre-release-gtk2/lib/libguile.so.17.1.1)
==4411==    by 0x1BF66152: scm_dapply (in /extra/paule/autobuild/Linux-kalypso-fc4-pre-release-gtk2/lib/libguile.so.17.1.1)
==4411==    by 0x1BF65656: deval (in /extra/paule/autobuild/Linux-kalypso-fc4-pre-release-gtk2/lib/libguile.so.17.1.1)
==4411==    by 0x1BF62DE3: deval (in /extra/paule/autobuild/Linux-kalypso-fc4-pre-release-gtk2/lib/libguile.so.17.1.1)
==4411==    by 0x1BF63AE3: deval (in /extra/paule/autobuild/Linux-kalypso-fc4-pre-release-gtk2/lib/libguile.so.17.1.1)
==4411==  Address 0x211EACE0 is 0 bytes inside a block of size 68 alloc'd
==4411==    at 0x1B90B5DA: operator new[](unsigned) (vg_replace_malloc.c:138)
==4411==    by 0x1C3A0BF2: CSuperpose::_superpose(CSSGraph*, CSSGraph*, int&) (ssm_superpose.cpp:2617)
==4411==    by 0x1C3A1194: CSuperpose::SuperposeCalphas(CSSGraph*, CSSGraph*, int*, int*, int, CMMDBManager*, CMMDBManager*, int, int) (ssm_superpose.cpp:2424)
==4411==    by 0x1C38F7C8: CSSMAlign::Align(CMMDBManager*, CMMDBManager*, int, int, int, int) (ssm_align.cpp:200)
==4411==    by 0x819179B: graphics_info_t::superpose_with_atom_selection(atom_selection_container_t, atom_selection_container_t, int, std::string, std::string, short) (graphics-info-superpose.cc:207)
==4411==    by 0x82483F3: superpose_with_atom_selection (c-interface-superpose.cc:201)
==4411==    by 0x1BE55150: _wrap_superpose_with_atom_selection(scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*) (coot_wrap_guile.cc:23440)





==4411== 30744 bytes in 10 blocks are definitely lost in loss record 452 of 509
==4411==    at 0x1B90B5DA: operator new[](unsigned) (vg_replace_malloc.c:138)
==4411==    by 0x1BBC6624: graphical_bonds_container::add_atom_centres(std::vector<coot::Cartesian, std::allocator<coot::Cartesian> > const&, std::vector<int, std::allocator<int> > const&) (Bond_lines.cc:2972)
==4411==    by 0x1BBC69E5: Bond_lines_container::make_graphical_bonds() const (Bond_lines.cc:1656)
==4411==    by 0x82D95EC: molecule_class_info_t::makebonds() (molecule-class-info.cc:2206)
==4411==    by 0x82E41D4: molecule_class_info_t::install_model(int, atom_selection_container_t, std::string const&, short) (molecule-class-info.cc:392)
==4411==    by 0x818C9FB: graphics_info_t::add_cb_to_terminal_res(atom_selection_container_t) (graphics-info-mutate.cc:76)
==4411==    by 0x81469DE: graphics_info_t::execute_add_terminal_residue(int, std::string const&, CResidue const*, std::string const&, std::string const&, short) (graphics-info-modelling.cc:1436)
==4411==    by 0x8209CD7: add_terminal_residue (c-interface-build.cc:528)


==4411== 8100 (7900 direct, 200 indirect) bytes in 25 blocks are definitely lost in loss record 438 of 509
==4411==    at 0x1B90B39A: operator new(unsigned) (vg_replace_malloc.c:132)
==4411==    by 0x1BD9E9C5: coot::ShelxIns::make_atom(coot::shelx_card_info_t const&, std::string const&, int, int, int, clipper::Cell&) const (coot-shelx-ins.cc:597)
==4411==    by 0x1BDA8730: coot::ShelxIns::read_file(std::string const&) (coot-shelx-ins.cc:375)
==4411==    by 0x832CDAF: molecule_class_info_t::read_shelx_ins_file(std::string const&) (molecule-class-info-other.cc:4453)
==4411==    by 0x820F502: read_shelx_ins_file (c-interface-build.cc:5983)
==4411==    by 0x81B8446: handle_read_draw_molecule_with_recentre (c-interface.cc:333)
==4411==



==4411== 82576 (880 direct, 81696 indirect) bytes in 11 blocks are definitely lost in loss record 193 of 509
==4411==    at 0x1B90B5DA: operator new[](unsigned) (vg_replace_malloc.c:138)
==4411==    by 0x1BBC6842: Bond_lines_container::make_graphical_bonds() const (Bond_lines.cc:1644)
==4411==    by 0x82D95EC: molecule_class_info_t::makebonds() (molecule-class-info.cc:2206)
==4411==    by 0x82E41D4: molecule_class_info_t::install_model(int, atom_selection_container_t, std::string const&, short) (molecule-class-info.cc:392)
==4411==    by 0x818C9FB: graphics_info_t::add_cb_to_terminal_res(atom_selection_container_t) (graphics-info-mutate.cc:76)
==4411==    by 0x81469DE: graphics_info_t::execute_add_terminal_residue(int, std::string const&, CResidue const*, std::string const&, std::string const&, short) (graphics-info-modelling.cc:1436)
==4411==    by 0x8209CD7: add_terminal_residue (c-interface-build.cc:528)
==4411==    by 0x1BE52918: _wrap_add_terminal_residue(scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*) (coot_wrap_guile.cc:19697)

----
Thu Mar 27 07:44:50 GMT 2008


    Frank: 

   "torsion general - crappy tail?"  I think I meant: why can't one
   use torsion general to swing a whole lot of protein around a main
   chain bond?  If I try, the mouse just clears after the 4th click.
   i.e., all I can move around is a miserable little tail of a
   sidechain, maybe.  Oh wait, I see now: yes, torsion general works
   only inside a residue.  Which is a bit anal, what if I want to
   swivel a whole domain or helix when it's changed from the MR model.
   Other place of nuisance: when modelling around in poor density at
   the end of a chain.

   "Ramachandran boundaries".  Ah yes: any reason why you don't have
   the same smooth boundaries that Richardsons use?  I've seen cases
   just outside your boundaries, but happily inside the Richardson
   one.

   "Views file: #Views".  Yes, the views file coot writes out itself
   doesn't load back.  Reason is (I remember) that the second (or
   third line) contains #Views, which screws up scheme.

   "Save coords: most recently changed".  Oh yes: could you populate
   the defaults (molecule; filename) based on the file most recently
   changed, and not on the one currently active?  These are not always
   the same, and chances are I want to save what I just built at, not
   what I was looking at.

----
Fri Mar 28 10:08:09 GMT 2008

    PRE:
    Use LSQ for NCS matrices.

----
Fri Mar 28 10:46:13 GMT 2008

  PRE:

     Just noticed another slightly weird thing: I mutated a couple of
     residues and added 3 in one NCS copy, then "NCS looped" them to
     the other subunit. The mutated ones in the second subunit had the
     right coordinates & atoms, but were still labelled with the old
     residue type.

     The inserted ones were on the end of the subunit

  Add tests for these.

----
Fri Mar 28 10:48:25 GMT 2008

    PRE:

      Somehow allow me to set that I want "Colour by Chain" on by
      default. (i.e. set-colour-by-chain for future molecules).
 
    Same for CA mode.

    (Pass default representation type to molecule_class_info_t's
    handle_read_draw_molecule?)

---
Fri Mar 28 10:51:31 GMT 2008

    CNS to coot: script needs a GUI?

----
Fri Mar 28 22:47:43 GMT 2008

Randy Read: 
Some time back I sent you a bug report about split-screen stereo.  Don't know 
if you've been able to reproduce it, but since then I've worked out the 
problems.  Basically, coot doesn't like the "Side by Side (Wall Eyed)" option 
any more.  Part of the problem is the saved state.  If the saved state is 
Mono or normal "Side by Side", then you're fine -- coot comes up the way you 
left it.  (Don't know about hardware, as I don't have access to an 
appropriate machine.)  If the saved state is "Side by Side (Wall Eyed)", then 
Draw -> Stereo... brings up a box claiming that you're still in that mode, 
but the view is a mono one.  The only way to get wall-eyed stereo from this 
state is to click "Mono" then "Side by Side" then finally "Side by Side (Wall 
Eyed)".  If you come up in Mono mode, you also can't go directly to 
wall-eyed -- first you have to go into normal side-by-side mode.


-----
Sat Mar 29 23:13:39 GMT 2008

    Documentation - including the functions should all be in one document
    (texi).

----
Sat Mar 29 23:15:20 GMT 2008

    update the density fit graph on delete residue (e.g. water).

---
Sun Apr  6 22:01:39 BST 2008

    Write a wiki script for munging/unmunging the atom names of NADP/H
    so that it can refine.

----
Mon Apr 14 09:33:36 BST 2008

Lothar Esser

I am using coot 0.5 pre 958 and noticed that after resizing the
window of the ramachandran plot and possibly zooming in (with the
right mouse) the scroll bars don't allow to pan over the plot. Only
under extreme zoom, the scroll bars allow a little bit of
adjustment. I have over 2000 residues and need to zoom and scroll a
bit if I want to click on residues.

I do get error messages on the
terminal that might help:

** CRITICAL **: file gtk-canvas.c: line 3518
** (gtk_canvas_set_pixels_per_unit): a ssertion `n >
** GTK_CANVAS_EPSILON' failed.

The same message repeats presumably every
time I scroll(?) or perphas zoom.

Also, I noticed after a long search that the -key zooms out after I
had zoomed in with the right-mouse. Could zooming out be added to the
mouse perhaps like Shift-right-mouse ?

Another feature of coot would be good: Show ramachandran plots by
chains that the user select i.e. a plot with one chain at a time.

---
Fri Apr 18 17:43:12 BST 2008

    There is a EM software wiki (scripps?) 

-----
Mon Apr 14 09:41:32 BST 2008

    can findligand be use to find just blobs (no input model?).  If
    not, make it so.

----
Mon Apr 14 09:42:28 BST 2008

    There is a server with uremediated files - from Kim? 
    Use this to validate new software.

-----

PISA: 

Buried surface area (average in pdb is about 40%).
	
Test for connected crystal

And min number of interfaces per chain (needs to be at lest 3 or the
crystal is not held in 3 dimensions)


---
Mon Apr 14 10:57:18 BST 2008

    add a function to display (a set of) spheres, 
    radius, position, colour

----
Mon Apr 14 11:29:17 BST 2008

    Jane Richardson

1rb1 bad clashscore.  embedded waters.  "Tassos' waters - from
ARP/warp

1k8z: geometry outlier

----
Mon Apr 14 11:43:59 BST 2008
    
  question to self: can you make a packing check quick?
  (I guess not).
    

----
Mon Apr 14 23:07:06 BST 2008

    chat with Nigel M.  
    Use CONNECT records in bonding (if possible)

-----
Tue Apr 15 14:02:16 BST 2008

How to run elal:

$ elal -pdb ~/data/rnase/tutorial-modern.pdb -map tmp.map -lib $warpbin/elal.lib

[there has to be a calculated map too, somewhere]

For checking carbohydrates:

wwww.glycosciences.de/tools/pdb-care


"ramachandran plots for carbohydrates"

however, need to do it for each linkage type.

----
Tue Apr 15 15:25:38 BST 2008
    
    Kim says Hursthouse for Portsmouth, Southhampton etc 
    small molecule crystal structures.  

* eCrystals
    
----
Wed Apr 16 11:23:49 BST 2008

   Add internal packing function? (bumps)

----
Wed Apr 16 11:36:44 BST 2008

    What is the RSR Z score from EDS?  
    Can we make something similar?

    Correlation?


----
Fri Apr 25 12:39:14 BST 2008

   When the startup script is not given as a dialog option but not
   accepted, then, on
   exiting, don't create a new state file.
    
    (better would be to give a GUI on exiting, shall we overwrite?)


---
Mon Apr 28 18:21:07 BST 2008

    
valgrinding migels rl shelx file:

(saving as pdb)

==12236== 621024 (187824 direct, 433200 indirect) bytes in 1092 blocks are definitely lost in loss record 302 of 312
==12236==    at 0x1B90B39A: operator new(unsigned) (vg_replace_malloc.c:132)
==12236==    by 0x1BD6A1B4: coot::util::deep_copy_this_residue(CResidue const*, std::string const&, short) (coot-coord-utils.cc:1414)
==12236==    by 0x1BDADCB8: coot::reshelx(CMMDBManager*) (coot-shelx-ins.cc:1961)
==12236==    by 0x1BDAE05F: coot::ShelxIns::write_ins_file_internal(CMMDBManager*, std::string const&) const (coot-shelx-ins.cc:997)
==12236==    by 0x1BDAF90F: coot::ShelxIns::write_ins_file(CMMDBManager*, std::string const&) (coot-shelx-ins.cc:988)
==12236==    by 0x8341182: molecule_class_info_t::write_shelx_ins_file(std::string const&) (molecule-class-info-other.cc:4437)
==12236==    by 0x8308DC6: molecule_class_info_t::save_coordinates(std::string) (molecule-class-info.cc:4228)
==12236==    by 0x81C8F50: save_coordinates (c-interface.cc:4042)
==12236==    by 0x81E3E45: save_coordinates_using_widget (c-interface-gui.cc:3248)
==12236==    by 0x82E83F7: on_save_coord_ok_button_clicked (callbacks.c:2190)
==12236==    by 0xAC4816: g_cclosure_marshal_VOID__VOID (in /usr/lib/libgobject-2.0.so.0.600.4)
==12236==    by 0xAB9284: g_closure_invoke (in /usr/lib/libgobject-2.0.so.0.600.4)

----
Thu May  1 20:08:33 BST 2008

    When waters are added with the W keybinding, let us know (with a
    dialog) if a "Pointer Atoms" molecule was created (may be pointer
    atom molecule was not set?)

----
Thu May  1 20:26:28 BST 2008

    1a02: place helix here at (9.2 36.6 28.3)

    1env: place helix here - many places.  Lots of errors.

    1xb7_sigmaa.mtz 2FOFCWT PH2FOFCWT (-16.5184307098389 50.2622299194336 19.386697769165)

       -> Helix overshoot


----
Thu May  1 20:55:01 EDT 2008

    Fixed atoms still fail - e.g. on TMC in 1e2k

-------

   When activating the restriants editor dialog, make the option menu 
   be the residue type at the centre of the screen.

   Priority: High
   Dificultly: Easy


-----

    A must before 1.0:

    revist add strand here, 
    check in low res maps.  
    Try other orientations of the cylinder
        score more badly atoms in negative density.

----
Sat May  3 11:33:52 EDT 2008

    Maria Chern:

    Can I have "view drag" on middle mouse, rather than ctrl left-mouse?

    Question to self: will that fix the single atom drag vs view drag
    problem that drives Erik insane?

-----
Sat May  3 12:46:24 EDT 2008

    c-interface-info.cc:3081 chiral volume restraint is checked
    against 7.  Should it be 6, because there no sigma is needed for a
    chiral volume restraint?

-----
Sat May 10 07:22:54 CST 2008

    Consider a test for wiggly ligands.

----

  For Sabine Schneider:

  How about "ideal geometry" restraints, that try to idealize torsion 
  angles to A or B form geometry?

-----

   How does side by side stereo work for maps when using (default on Mac) 
   (set-display-lists-for-maps 1)

-----

   Do NCS map average. then denisty fit analysis vs the average map -
   the scores go down vs the first map (adding not averaging suggests
   eleanor?)

   The NCS average *is* being averaged.  It could a result of the
   density outsite the averaged block being set to zero.

----

   look at graphics-info-t::execute_add_terminal_residue() to see
   masking a map by atoms around a point.

----

   ask Stuart about line 113 in mmdb_chain.h
   virtual pstr GetEntryID()

   line 133 mmdb_symop.cpp CSymOp::GetSymOp()


----
Tue Jun  3 16:19:39 BST 2008
    
    valgrinding a greg test run:


==13600== 3 errors in context 55 of 212:
==13600== Invalid read of size 1
==13600==    at 0x42048AE: coot::protein_geometry::try_dynamic_add(std::string const&, int) (protein-geometry.cc:2085)
==13600==    by 0x42052F2: coot::protein_geometry::have_dictionary_for_residue_type(std::string const&, int) (protein-geometry.cc:2408)
==13600==    by 0x80FEDAC: graphics_info_t::check_dictionary_for_residues(CResidue**, int) (graphics-info-modelling.cc:484)
==13600==    by 0x810463A: graphics_info_t::copy_mol_and_refine(int, int, int, std::string, int, std::string, std::string, std::string) (graphics-info-modelling.cc:265)
==13600==    by 0x8106882: graphics_info_t::copy_mol_and_regularize(int, int, std::string, int, std::string, std::string, std::string) (graphics-info-modelling.cc:107)
==13600==    by 0x8106CF7: graphics_info_t::regularize(int, short, int, int) (graphics-info-modelling.cc:688)
==13600==    by 0x819FF5E: regularize_zone (c-interface-build.cc:5915)
==13600==    by 0x44391A0: _wrap_regularize_zone(scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*) (coot_wrap_guile.cc:14018)
==13600==    by 0x4559B90: scm_gsubr_apply (gsubr.c:221)
==13600==    by 0x4540523: scm_dapply (eval.c:4915)
==13600==    by 0x4542724: deval (eval.c:4364)
==13600==    by 0x4543A20: deval (eval.c:3384)
==13600==  Address 0x6145E14 is 12 bytes inside a block of size 14 free'd
==13600==    at 0x4023096: operator delete(void*) (vg_replace_malloc.c:244)
==13600==    by 0x56B890C: std::string::_Rep::_M_destroy(std::allocator<char> const&) (in /usr/lib/libstdc++.so.6.0.9)
==13600==    by 0x56BA7B0: std::string::~string() (in /usr/lib/libstdc++.so.6.0.9)
==13600==    by 0x42048AD: coot::protein_geometry::try_dynamic_add(std::string const&, int) (protein-geometry.cc:2084)
==13600==    by 0x42052F2: coot::protein_geometry::have_dictionary_for_residue_type(std::string const&, int) (protein-geometry.cc:2408)
==13600==    by 0x80FEDAC: graphics_info_t::check_dictionary_for_residues(CResidue**, int) (graphics-info-modelling.cc:484)
==13600==    by 0x810463A: graphics_info_t::copy_mol_and_refine(int, int, int, std::string, int, std::string, std::string, std::string) (graphics-info-modelling.cc:265)
==13600==    by 0x8106882: graphics_info_t::copy_mol_and_regularize(int, int, std::string, int, std::string, std::string, std::string) (graphics-info-modelling.cc:107)
==13600==    by 0x8106CF7: graphics_info_t::regularize(int, short, int, int) (graphics-info-modelling.cc:688)
==13600==    by 0x819FF5E: regularize_zone (c-interface-build.cc:5915)
==13600==    by 0x44391A0: _wrap_regularize_zone(scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*, scm_unused_struct*) (coot_wrap_guile.cc:14018)
==13600==    by 0x4559B90: scm_gsubr_apply (gsubr.c:221)


==13600== 
==13600== 1 errors in context 5 of 212:
==13600== Invalid read of size 8
==13600==    at 0x828797E: molecule_class_info_t::closest_atom(coot::Cartesian const&, bool) const (molecule-class-info.cc:330)
==13600==    by 0x8287AC2: molecule_class_info_t::closest_atom(coot::Cartesian const&) const (molecule-class-info.cc:313)
==13600==    by 0x81877BE: active_atom_spec() (c-interface-info.cc:1108)
==13600==    by 0x818E856: active_residue() (c-interface-info.cc:1139)
==13600==    by 0x442DD86: _wrap_active_residue() (coot_wrap_guile.cc:29092)
==13600==    by 0x4542A78: deval (eval.c:4122)
==13600==    by 0x4544012: deval (eval.c:3635)
==13600==    by 0x4544012: deval (eval.c:3635)
==13600==    by 0x45406C6: scm_dapply (eval.c:4997)
==13600==    by 0x45408B2: scm_apply (eval.c:4796)
==13600==    by 0x45454DC: scm_call_0 (eval.c:4651)
==13600==    by 0x45A2F1F: scm_body_thunk (throw.c:356)
==13600==  Address 0x5BBCB78 is not stack'd, malloc'd or (recently) free'd
==13600== 


==13600== 1 errors in context 6 of 212:
==13600== Conditional jump or move depends on uninitialised value(s)
==13600==    at 0x80E8E1A: clipper::Map_stats::Map_stats<clipper::Xmap<float> >(clipper::Xmap<float> const&) (map_utils.h:123)
==13600==    by 0x40465EA: coot::ligand::find_clusters_internal(float, std::vector<clipper::Coord_orth, std::allocator<clipper::Coord_orth> > const&) (ligand.cc:774)
==13600==    by 0x40471E1: coot::ligand::find_clusters(float) (ligand.cc:762)
==13600==    by 0x81D2E42: execute_ligand_search_internal() (c-interface-ligands.cc:953)
==13600==    by 0x81D3A11: execute_ligand_search (c-interface-ligands.cc:854)
==13600==    by 0x4435F16: _wrap_execute_ligand_search() (coot_wrap_guile.cc:18153)
==13600==    by 0x4542A78: deval (eval.c:4122)
==13600==    by 0x4544012: deval (eval.c:3635)
==13600==    by 0x4544012: deval (eval.c:3635)
==13600==    by 0x45406C6: scm_dapply (eval.c:4997)
==13600==    by 0x45408B2: scm_apply (eval.c:4796)
==13600==    by 0x45454DC: scm_call_0 (eval.c:4651)


==13600== 
==13600== 1 errors in context 1 of 212:
==13600== Invalid read of size 4
==13600==    at 0x49FF2A9: CResidue::GetAtomTable(CAtom**&, int&) (mmdb_atom.cpp:2568)
==13600==    by 0x82879DC: molecule_class_info_t::closest_atom(coot::Cartesian const&, bool) const (molecule-class-info.cc:343)
==13600==    by 0x8287AC2: molecule_class_info_t::closest_atom(coot::Cartesian const&) const (molecule-class-info.cc:313)
==13600==    by 0x81877BE: active_atom_spec() (c-interface-info.cc:1108)
==13600==    by 0x818E856: active_residue() (c-interface-info.cc:1139)
==13600==    by 0x442DD86: _wrap_active_residue() (coot_wrap_guile.cc:29092)
==13600==    by 0x4542A78: deval (eval.c:4122)
==13600==    by 0x4544012: deval (eval.c:3635)
==13600==    by 0x4544012: deval (eval.c:3635)
==13600==    by 0x45406C6: scm_dapply (eval.c:4997)
==13600==    by 0x45408B2: scm_apply (eval.c:4796)
==13600==    by 0x45454DC: scm_call_0 (eval.c:4651)
==13600==  Address 0x3E10912E is not stack'd, malloc'd or (recently) free'd
==13600== 

---
Tue Jun  3 17:14:33 BST 2008

    Make a new source file with the widget work in c-interface-info.cc
    c-interface-info-gui.cc


----

    JB:

    Refine by sphere. IMPORTANT!

    Render Geometry objects in raster3d

    For beauty:
      Consider putting a std::vector<coot::generic_display_object_t>
      generic_objects into raytrace_info_t, to store info for generic
      objects in ray trace info.


----


    check the number of saves on Eigen flips.  On undoing, it seems I
    have to undo twice for each step.

    Priority: Critical.

    
----
Thu Jun  5 16:17:38 BST 2008

    Conversation with Robert von Dreele:

    Create a "What's New in Coot" button - which gets the latest
    Release Notes from the web server and displays it in a text
    widget.

    Should be done in a sub-thread


----
Sat Jun  7 03:45:44 BST 2008

MolProbity: all-atom contacts and structure validation for proteins
and nucleic acids

  Ian W. Davis1, Andrew Leaver-Fay2, Vincent B. Chen1, Jeremy
  N. Block1, Gary J. Kapral1, Xueyi Wang2, Laura W. Murray1, W. Bryan
  Arendall, III1, Jack Snoeyink2, Jane S. Richardson1 and David
  C. Richardson1,*

  Sugar puckers

     RNA sugar pucker (C3' endo or C2' endo) is strongly correlated to
     the perpendicular distance between the following (3') phosphate
     and either the plane of the base or the C1'-N1/9 glycosidic bond
     vector. Incorrectly chosen sugar puckers also often result in
     out-of-range values for the epsilon dihedral. This is important
     information, because a sugar pucker is very difficult to
     determine directly from the electron density at resolutions
     typical for large RNAs. MolProbity checks epsilon angles and
     checks the modeled sugar pucker against the base-phosphate
     distance; it flags outliers as potentially having the wrong
     pucker.

  Nucleic Acids Research, 2007, Vol. 35, No. suppl_2 W375-W383

------
Sun Jun 15 15:40:10 BST 2008

    Activate readline in guile build (at least for my computers!)


------
Wed Jun 18 02:51:47 BST 2008

    "Mark all for Close" button in the close molecule dialog.

----
Wed Jun 18 12:46:02 BST 2008

    Distances and angles dialog is a transient(?) and needs to be
    position-saved like the other transients.

-----
Wed Jun 18 22:33:25 BST 2008

    Fun: How about a script to make ray traced images of ligands at
    active sites.  Use kinases like these (find others that match but
    have different ligands).  Need to find structures with data of
    course.  Consider using j-multi-read, get the matrices and apply
    them to their maps.  Maybe that is not needed, because we are not
    going to look at the overlapped maps anyway, just ligands in the
    binding sites.

    Data needs to be formatted differently in the
    text control file - we need a specification of the mtz file and
    column labels too.

coot-download/3bv2.pdb
coot-download/2zb0.pdb
coot-download/3Coi.pdb

-----
Wed Jun 18 22:40:45 BST 2008

   RAPPER notes:

   Loop modelling with section not currently modelled in input PDB:

rapper params.xml model-loops --pdb "test.pdb" --map "test.map" --chain-id "B" --models "1" --cryst-d-high "1.77" -seq ~AAA~ --start 705 -stop 708 --use-CCP4i-file-name true --ccp4i-file-name "rapper_1" --runs-dir "test_dir" --edm-fit true --sidechain-mode smart --sidechain-radius-reduction 0.75 --sidechain-library RAPPER-DIR/data/richardson.lib 

   Loop modelling with section already modelling in input PDB file using electron density:

rapper params.xml model-loops --pdb "test.pdb" --map "test.map" --chain-id "B" --models "1" --cryst-d-high "1.77" --start 705 -stop 708 --use-CCP4i-file-name true --ccp4i-file-name "rapper_1" --runs-dir "test_dir" --edm-fit true --sidechain-mode smart --sidechain-radius-reduction 0.75 --sidechain-library RAPPER-DIR/data/richardson.lib 

-----
Thu Jun 19 09:13:48 BST 2008



==27246== Invalid read of size 1
==27246==    at 0x4007203: strlen (mc_replace_strmem.c:242)
==27246==    by 0x709A16A: std::string::string(char const*, std::allocator<char> const&) (in /usr/lib/libstdc++.so.6.0.10)
==27246==    by 0x8178864: make_directory_maybe (c-interface.cc:7080)
==27246==    by 0x80FA10C: c_inner_main (c-inner-main.c:246)
==27246==    by 0x410392C: invoke_main_func (in /home/juditd/coot-build-3/autobuild-installed-pre-release-gtk2/lib/libguile.so.17.1.2)
==27246==    by 0x40D6409: c_body (in /home/juditd/coot-build-3/autobuild-installed-pre-release-gtk2/lib/libguile.so.17.1.2)



----
Sat Jun 21 01:55:45 BST 2008

    Joel Bard:

    When coming out of side-by-side stereo, the map disappears -
    something to do with the GL context the display list is generated
    in?

    Priority: Medium/Low  (fix refine by sphere first)

    

----
Sat Jun 21 12:39:55 BST 2008


   "Triple refine" (manual-refine-residues 1) at the C terminus should refine
   2 residues (like autozone does, I think).


----
Sat Jun 21 14:56:11 BST 2008

running loopy:

$flex_wARP_warpbin/loop -paramfile=rnase_2_1_com.tmp

---
Wed Jun 25 16:40:23 BST 2008

    a function similar to shell-command-to-string, but that takes
    command line args and data lines and returns a list of strings.

    Use it in get-refmac-version.


---
Mon Jun 30 15:20:48 BST 2008

    asc_to_graphics() and all of atom-utils can go (need to fix up
    skeleton/BuildCas.cc)
    
    For 0.6


-----
Wed Jul  2 08:27:50 BST 2008

    Do rotamers. (55B)

    Do Residue Info.  Intermediate atoms disappear.  Why?

----
> > In message <B0848263-CEEB-48EF-B991-4ED71F81E9D4@idi.harvard.edu>  
> > Jianghai Zhu
> > <jzhu@idi.harvard.edu> writes:
> >> 
> >> I am using the nightly build of coot prerelease.  I found out that
> >> every time when I switch the stereo on and off, I lose the electron
> >> density map.  If I scroll the mouse, I can bring the map back.  But  
> >> if
> >> I have multiple maps on display, then only one map can be brought  
> >> back.

----
Thu Jul  3 12:44:23 BST 2008


    on_draw_hydrogens_yes_radiobutton_toggled 

    This draw hydrogens thing should be a single check button, not 2
    radio toggle buttons.

    Get rid of the "Apply" button.

    Low priority


----
Mon 14 Jul 02:17:01 2008 BST


   molecule_class_info_t::handle_read_draw_molecule() has:

      // update the maps so that they appear around the new centre. 
      // 
      if (reset_rotation_centre) 
	 for (int ii=0; ii<g.n_molecules(); ii++) {
	    g.molecules[ii].update_map(); 
	 }

   It should not be there.  That should be outside of the
   molecule_class_info_t class.


-----

    How shall I fix the ramachandran restraints?

I need to investigate grad.tab when I try with a triple in the
graphics.

A hidden scripting interface to turn on numerical gradients?  Put that
in graphics-info-modelling and copy_mol_and_refine() after we
make_restraints().


----
Mon Jul 14 15:23:21 BST 2008

    revision 1274: before (big) messing with torsion restraints

---
Mon Jul 14 23:20:37 BST 2008

    Phil Evans:

    new waters are marked as HETATMs?  They should be.

----
Thu Jul 17 13:43:34 BST 2008

in void
on_refine_params_use_torsions_checkbutton_toggled (GtkToggleButton *togglebutton, 
						   gpointer         user_data)

do we need to unset the Rama angle refinement?  (and toggle the box too?)


----
Thu Jul 17 23:33:32 BST 200


int
coot::restraints_container_t::make_flanking_atoms_rama_restraints(const protein_geometry &geom)

	 // super careful would mean that we check the link type of
	 // both pairs before calling this function:


---
Fri Jul 18 12:33:10 BST 2008

   Penelope build-hardy:

   0-coot.state.scm-fails-to-reset-view

---
Fri Jul 18 14:55:30 BST 2008

    Get a residue.  do b factor analysis

    copy the residue. do b factor analysis.

    delete some atoms from mol 1.  the wrong plot updates.

----

to get what_check output for all pdb entries

www.cmbi.ru.nl/pdbreport/ab/1abc/pdbout.txt


-----
Tue Jul 22 16:06:11 BST 2008

    Robbie J suggests:

    Check and fix nomenclature errors for ARG.

    Put Fix nomenclature errors into the GUI.

    Schedule this for 0.6.


------
Wed Jul 23 12:28:06 BST 2008

    Think about adding Thomas' Phi/Psi (energy) plots (of the torsions
    around the glycosidic linkage) and plotting the points for the
    linked carbodydrates in the molecule.


----
Thu Jul 24 10:40:06 BST 2008

    1mn8 D52 do-180-degree-side-chain-flip must work! 
    chi5 must be recognised.


------

   a BMA in 3d2u: trying to refine a BMA, should Coot on the fly
   generate the BMA full dict from the minimal description?


-----
Mon Jul 21 12:33:21 BST 2008


in molecule_class_info_t::handle_read_draw_molecule():


      // update the maps so that they appear around the new centre. 
      // 
      if (reset_rotation_centre) 
	 for (int ii=0; ii<g.n_molecules(); ii++) {
	    g.molecules[ii].update_map(); 
	 }

    
    and symmetry likewise.

    Bleugh - needs a fix.  Needs to be in
    handle_read_draw_molecule_with_recentre in the c-interface
    function


---
Mon Jul 28 23:36:45 BST 2008

    function to strip spaces from a string (scheme).

----
Mon Jul 28 23:39:48 BST 2008

    rework const char *atom_info_string(int imol, const char *chain_id, int resno,
			     const char *ins_code, const char *atname,
			     const char *altconf) 

    to return a SCM.  And test that it gives #f when atom not exist.



-----
Tue Aug  5 23:11:18 BST 2008

    Here's another rama test.

    Turn on Rama restraints. turn on helical restraints.  try to turn
    residues 84 to 93 into a helix.  (you can cut them out from the
    rest of the protein to make things clearer).

----
Wed Aug 13 12:44:28 BST 2008

    It would be nice if the scripting window(s) were available via
    some sort of (iconfied?) combo box in the horizontal toolbar.

    rajan@mbu.iisc.ernet.in 12 Aug 2008.

----
Wed Aug 13 12:52:23 BST 2008

    Fix up the atom ordering in modified (added to) residues
    Steven Sheriff 12 Aug 2008.


---
Thu Aug 14 23:18:24 BST 2008

    EJD: I want to do an atom selection for my Ramachandran
    Plot. (e.g. I want to see just one chain)


---
Mon Aug 18 16:15:01 BST 2008

    How about a "fast save" that saves pdb files with modifications,
    and the state file... bound to a key.


----
Fri Aug 22 18:24:20 BST 2008

    IUCr: make a GUI interface to mutate-by-overlap.



-----
Tue Sep  2 09:53:52 JST 2008

    what happens when you try to rigid body refine an a fragment from
    ribosome (or something big)?

    If utter slowness, consider introducing a function to turn off
    fixed-atom masking.


---
Wed Sep 10 13:43:57 BST 2008

    When rotating atom in chi or rotamers, change the aniso B values too.

---
Thu Sep 11 07:18:38 BST 2008

    Reasons why the builds fail:

    o because the sourceforge host for the gtkglext package disappeared
      (dynamic/direct hostname)

    o Ran out of memory, causing the compiler to crash (I'd forgotten
      to add swap)

    o Forgotten to add new files to the svn repo


---
Sun Sep 14 09:49:35 BST 2008

    Failures/Issues at the BCA Summer School

   o Recover session didn't work - "Nothing to Recover".

   o "Add Strand here" failed - dictionary of structures, either
     didn't exist or was in the wrong place: 
     C:\share\coot\ss-reference-structure..  found 0 pdb files

   o Rotamer probability tables not found - falling back to simple 
     (confusing) rotamer probabilities
     C:\msys\home\bernhard\autobuild\MINGW..\share\coot\rama-.. not found

   o Refmac button failed

----
Sun Sep 14 10:45:38 BST 2008

    There is a dialog for globularize protein.  It needs to be
    activated. (Some work).

----
Sun Sep 14 12:32:23 BST 2008

    Note: the right way to set an active menu item for an option menu
    is in fill_option_menu_with_undo_options().


----
Tue Sep 16 12:56:16 BST 2008

    Note: Processes for automated builds/testing

 o The process that checks out the latest sources, compiles the code
   and publishes a source tar file, the build log (for building
   problems) and build-from-source success status

 o The build+test+tar+(copies+)publish process itself - build-it, runs
   on all up to hosts several times a day

 o The process that cleans up the build directory - every day, deletes
   everthing more than 7 day old

 o The process that deletes "old" pre-release build tar files from the
   web - runs every day, deleting everything more than 30 days old

 o The process that makes a web page from the build results - runs
   every 5 minutes, examining results from fedrated builds


---
Tue Sep 16 13:22:03 BST 2008

    key bindings

* Done

    non-clickable molecule

* Done (apparently)

    molprobity dots problem

* Done.

    render graphical objects

* Done.


-----
Thu Sep 11 15:43:22 BST 2008

    Warn when making a screenshot of a density map that is dark blue.

----
Thu Sep 18 13:01:38 BST 2008

    Add text to Raster3D input.r3d

----
Mon Sep 22 17:53:14 BST 2008

    "Cocktail test this blob" vs 200 candidates (to be addeded to
    distro).  Allow the user to add their own ligands.

    Medium to high importance.


----
Tue Sep 23 09:52:40 BST 2008

[Note to self, rigid body fit is not working/did not complete? after
ligand fitting, should fit against masked map (the one that gets
exported to graphics after ligand fitting has completed].

[Note to self, make a set of slides for these]

----

Mon Oct  6 13:56:30 BST 2008
PRE RELEASE CHECKS

Does ligand fitting work?
Does torsion bond hilighting work?

----
Wed Oct  1 14:14:26 CEST 2008

   What is Wadley et al. RNA torsion angles? P and C4'

   Mentioned by Tim Grune

----
Fri Oct  3 15:23:51 CEST 2008

  o Note: Randy says that the maps from phenix autobuild don't work -
  all red in the validation (traffic lights?) - I guess that the maps
  are scaled up.

----

   For a tutorial: design a drug in libcheck/sketcher and fit it to
   the density - use something from the PDB.

* How about APN/aspirin?

---
Mon Oct  6 13:56:44 BST 2008
   
	It's not the algorithms

   After the release, there are no pre-release source tars, so the
   build web page builder fails (a list-ref of a #f list)

   Python 2.5.2 build tests are failing with sendmail on FC5 (after
   re-install?)

   Crash on CentOS 4 machines when trying to read svgs

   Trying to separate python build logs from non-python build
   logs. Problems! overlapping dependencies, test the tar file and
   transfer it - many fixes to sort out.

   Coot builds on 32 and 64 bit systems.  In which directory should
   the libraries go?  By convention, 32 bit libs in prefix/lib and 64
   bit libs in prefix/lib64.  How is this sorted out in the configure?
   Buy using an autoconf macro AC_LIB_PREPARE_MULTILIB.  Where is this
   provided? In a recent version of gettext (more recent that fink
   currently provides).  So on every build system we have to get (e.g
   from the pre-build package e.g. Ubuntu intrepid) or downlaod and
   install gettext manually and point the autogen (that run aclocal and
   autoconf) to use the macros in the files installed from gettext.

   I wanted to check that the torsions in ligand conformation were
   being correctly set.  I needed a function therefore to return a
   torsion to the scripting level - given a set of atom specifiers.
   So first a wrote a test that tested that the torsion generation
   function worked - did it fail properly when the (some of) the atoms
   were not correctly specified (i.e. not in the molecule).  When that
   was working, I could write a test to do the ligand fitting and
   check the torsions of the best fitting (hopefully correctly
   fitting) model.

   

---- Wed Oct 8 14:11:40 BST 2008

   coot-shelx-ins.cc
   // 20081008 allow for a site occupancy factor on the PART line
   // but then throw it away.  Not good. Should fix at some stage.

---
Fri Oct 17 13:43:26 BST 2008

   The crash catcher should check that the core belongs to coot before
   giving dialog with text. e.g. it fails to read the correct core on
   jackal (run from ~).

---
Fri Oct 17 14:13:50 BST 2008

   main.cc: 
   // else we need to pass cld.alternate_splash_screen_file_name
   // to a function that accepts a file_name a makes a splash
   // screen.  Hmm..

----
Thu Oct 23 11:54:39 BST 2008
  
   It would be nice if baton building would orient as it translates 
   so that the baton is along the screen Y.

----
Sat Oct 25 03:11:07 BST 2008

  In the conformer generation, generate many models first, then
  select the top N for energy minimization.

  Instead of search the maps for ligand sites, try to position the
  ligand at a give postion (e.g. the screen centre). 

  Try a 27 pt box search around the centre position.  Try different
  rigid body angles at the box points.


----
Tue Oct 28 20:14:17 GMT 2008

  on reading the ribosome structure, 99% of 110 - (about 2) seconds is
  spend here:

	MMDBManager->ReadCoorFile(pdb_name.c_str());

---
Wed Oct 29 08:16:59 GMT 2008

   Frank:

     This one is quicker: are you able to colour text in the Display
     Manager?  In which case it would be handy to have the text for
     the map objects coloured according to map type, e.g. blue for
     sigmaA map, red for difference map (info you can infer from the
     mtz?), and even green for other maps (usually
     experimental/averaged).

   Yes, except use the background color, I think.  And how about a
   button that is half red, half green?


---
Wed Nov  5 17:47:55 GMT 2008

    For restraints editor hilighting: generate graphics objects

    restraints_editor_hilight_bond(residue_type, atom_name_1, atom_name_2)
    restraints_editor_hilight_angle(residue_type, atom_name_1, atom_name_2, atom_3)
    restraints_editor_hilight_atom(residue_type, atom_name_1)
           draw a ring around the atom say.  Maybe in 2 planes

    torsion and plane are a lot more difficult


---
Fri Nov  7 08:07:07 GMT 2008

    3bmw: difference map 11, near 645 ILE, surely there is something
    wrong with the "peak next to molecule" algorithm?

----
Sun Nov 16 16:42:42 GMT 2008

   look at xnee for testing the gui (needs linux).

----
Sun Nov 16 16:46:20 GMT 2008

   run test_yourself() for rotamer_probability_tables needs to be
   added to test suite.


---
Sat Nov 29 15:08:44 GMT 2008

   In make_link_restraints() we need to make sure that we are sensibly
   setting the fixed atom/residue flag.
	
   Where is the test that these residues are L-peptides? in
   add_rama()?  No.  So it should be in make_rama_triples() then.
   Perhaps.  But it should be somewhere.

---
Thu Dec  4 16:27:58 GMT 2008

   Conversation with Ed:
 
  o Nice to have ^S bound to Save Coords (Add accelerator to gui)

  o Can you generate "arrow key-press events" from rotating the
  PowerMate?  Surely yes, it's just a matter of getting the power mate
  in the mode to generate the characters.  Ah.. but no, the powermate
  can generate 3 key presses basically (in Bills example, F1, F2, F3)

   Hmm...

  o What about binding PowerMate to chi angle motion - which is moving
    the mouse a bit.  Surely we can do that?

* Yes we can.

---
Sun Dec  7 19:36:04 GMT 2008

   simple-restraint.hh

	    // 20081207: don't do this here now - because the
	    // memory/selection is deleted again in
	    // clear_up_moving_atoms(). It *should* be done here of
	    // course, but we'll save that for the future.
	    // 
	    //if (atom) { 
	    // mol->DeleteSelection(SelHnd_atom);
	    // atom = NULL;
	    // } 

   NOTE: In future problems of memory when using a CRresidue * vector
   could be due to restraints_container_t deconstuctor.

---
Wed Dec 10 10:16:05 GMT 2008

	Add Use of 1hvv NCS residue range maps to advanced tutorial.

---
Thu Dec 11 12:55:32 GMT 2008

   After an mmdb read, delete chains that don't have atoms in them.
   (created by LINK? records in the header).

---
Fri Dec 12 00:01:19 GMT 2008

   When writing out the state file (header?) only include pdb
   molecules that can be read.

   e.g. after using copy fragment, but no saving of the molecule.

----
Fri Dec 12 00:07:18 GMT 2008

   Tutorial II ideas: 

   Find and fix the bad trans bond in 1h4p

---
Fri Dec 12 13:23:52 GMT 2008

   Roberto Steiner wants CA + ligs + waters mode.


---
Mon Dec 15 14:07:50 GMT 2008

INFO::
	How to do a release
	How to make a release

   o New entry in ChangeLog
   o Make sure RELEASE-NOTES are up to date and add a ----- marker at the top
     Make sure the release is called Release, not pre-release
   o change the version in configure.ac
   o artists acknowledged in glade/aboutdialog?
   o change the splash screen
        Use glade-2
        Fix the release name in the About dialog (aboutdialog)
        and copyright years.
        To correct the transparency, duplicate layer, bucket fill with white 
        the bottom layer, merge and bevel 6.
   o git add the new splash screen pixmap
   o Change the pixmaps/Makefile.am for the current splash screen image
   o git commit -a
   o git push [maybe here? not sure]
   o git tag Release-0.x.y
   o git commit
   o git push origin Release-0.x.y  # publish the tag
   o Passes build (build-latest) on pc?
        OK, disable the build-latest cron-job 
   o clear out pre-releases
   o check for updates - needs consideration?
   o Build binaries on some hosts
   o Update latest version info on coot web page, Projects/coot/web/index.html
   o Mail mailing list

---
Mon Jan 12 00:17:05 GMT 2009

   INFO:: run libcheck from a pdb file:

Y
_FILE_L /home/paule/ccp4i/rnase/STU_bondlist.cif
_MON STU
_SRCH N
_FILE_O /home/paule/ccp4_scr/RNAse_22
_REF
_END

Fine but where did /home/paule/ccp4i/rnase/STU_bondlist.cif come from?


---
Mon Jan 12 17:26:40 GMT 2009


   INFO::

   Nice example of sphere refinement:

   set 37 B (or A) to rotamer 1.  

   Do "r" refinement.  -> Wrong solution

   Do "R" (sphere) refinment, PHE gets kicked and goes to the right
   solution (I wonder if refmac would do that?)


---
Mon Jan 12 22:43:50 GMT 2009

	in wiggly-ligands, add an option to invert chiral centres
	(just in case the chemist made the wrong thing).


----
Tue 20 Jan 2009 10:13:19 EST

    Heres a failure that should not be...

    Get monomer -> PAD.
 
    Delete residue 3 from 1a72 and try to fit libchecked PAD.


-----
Fri 23 Jan 2009 17:31:50 EST

  Paul Hubbard:  I want to do rigid body refinement on the fragment
   of the active residue - don't make me click 2 residues to define
   the fragment - find it yourself 

* You can do this now with "X" key binding (laertes)


----
Sun 25 Jan 2009 04:57:33 EST
 
   How about adding the rmsd bonds, angles (etc?) along with the 
   chi squareds. 


----
Tue Feb  3 14:55:32 GMT 2009

   INFO:::

	std::FILE (e.g. used by gmp.h) is resolved by including cstdio
	(e.g. before including libguile.h).

----
Wed Mar  4 13:12:21 GMT 2009

    Metal distances: 

    http://tanna.bch.ed.ac.uk/

-----
Sat Feb 28 19:36:50 GMT 2009

	when testing new tree code:

     test: torsion general
     test: wiggly ligand best solution C* N9 C1* C2* is correct
     test: var4 and var4 for ATP and var1s for MPR


     Note: revision 1917 has "new" mgtree code, which seems to work
     for wiggly ligands, but torsion angles are horrifically broken.


----
Thu Mar  5 10:26:43 GMT 2009

    Torsion general should optionally wag the chain or molecule (not
    just the rest of the residue).  Frank von Delft, Cathy Lawson.
    Maia Cheney.


-----

   Better Credit for USF in get pdb and data.

   3f5p has a nice ligand


-----
Tue Mar 24 19:05:54 GMT 2009

     Let the Display Manager auto-resize on reading new pdb or map.
	[Frank von Delft]


---
Wed Mar 25 19:09:38 GMT 2009

    A tool to split out ions/ NA/ CL from a particular chain(?) and
    put them in their own chain.

    Then isSolventChain() can work, and so renumber-waters can work.


----

     Add a residue/chain selector (an mmdb string entry) to allow the 
     selction for a ramachandran plot (put it at the bottom of the dialog).


-----
Thu Mar 26 18:39:42 GMT 2009
   
     Adding a terminal residue: 

     the phi/psi pairs from a ramamchandran plot can't be the phi/psi
     for *one* residue, they
     must be psi of one and phi of the next (or the other way round).
     Hmm... needs some though.
     

---
Fri Apr 17 17:38:02 BST 2009

	 Worry about the generation of coot_wrap_*_pre.cc.  Copying
	 them from the build directory back to source dir.  Do it more
	 often so that these files are not out of date.

	How can this be made easier/automated?


---
Sat 16 May 2009 12:40:38 BST

	It would be nice if when batton building, Coot realized that
	the maps was not sufficiently smoothly sampled and suggested
	to resample it.


----
Thu May 21 15:11:17 BST 2009

    Ligand search:

    Fit libcheck_ZMR.pdb (jackal) to 2cml (without A1477 of course)

    "Right here" mode.


----
Fri May 15 14:37:13 BST 2009

   how to move an svn repo:
http://whynotwiki.com/How_I_moved_my_code_repository_to_Google_Code


   how to svn revert a particular revision (just this one) 
   $ svn merge -c -5702 . 

--
Sun 17 May 2009 13:01:52 BST

compiling on bragg3

c-interface-mmdb.cc: In function 'CMMDBManager* mmdb_manager_from_python_expression(PyObject*)':
c-interface-mmdb.cc:326: warning: deprecated conversion from string constant to 'char*'
c-interface-mmdb.cc: In function 'PyObject* display_python(PyObject*)':
c-interface-mmdb.cc:424: warning: deprecated conversion from string constant to 'char*'


----
Sun 24 May 2009 11:27:12 BST

	gtk2-build-status on cycle seems to being written by a gtk2
	build.  

----
Mon May 18 23:28:41 BST 2009

	Add to GUI talk: What's the difference between:
        Some function
	and 
	Some function...
       	?

----
Wed May 20 00:16:05 BST 2009

	Have go to atom open... on a bulding structure
	
	file -> get 1gxt from pdb 
	
	(change molecule to (new) 1gxt
	
	but the chains in the residue/chain tree do not change!

	Bug!


---
Thu 28 May 2009 23:16:29 EDT

   Bug!

   (On mac) restraints editor -> edit ESD for O C in ACT -> 0.0 (or
   anything) -> Coot resets it to 0.02 Grr.


---
Sun 31 May 2009 09:32:42 EDT

	read-cif-dictionary should return a result and should be
	tested against in greg tests.


---
Tue Jun  2 02:52:49 EDT 2009

   Herb Klei:

   unsaved changes dialog should have a check button that
   allows/dissallows showing unsaved-only/all as menu items in the
   option menu.

   It should start up as default showing only unsaved molecules.
  
   See graphics_info_t::fill_unsaved_changes_dialog().


* Done - sort of.  The active molecule is the first unsaved molecule.
  It would be better to have a check button here:
  "Just molecules with unsaved changes?"

---
Mon Jun  8 17:28:43 BST 2009


	on rama refining pdb1h3y.ent (at the end, I think)

In /home/paule/autobuild/Linux-kalypso-pre-release-gtk2/share/guile/gtk-2.0/gtk.scm:
 147:  0* [apply #<procedure #f ()> ()]
In unknown file:
   ?:  1  [#<procedure #f ()>]
In /home/paule/autobuild/Linux-kalypso-pre-release-gtk2/share/coot/scheme/fitting.scm:
    ...
 212:  2  (let* ((imol-map #)) (func (car *multi-refine-spec-list*) imol-map) ...)
 213:  3* [fit-protein-rama-fit-function (4 "B" -10000 #f) 5]
 173:  4  (let ((imol #) (chain-id #) (res-no #) ...) (let (# #) (map # #) ...))
 177:  5  (let (# #) (map # #) ...)
 179:  6* [map #<procedure #f (alt-conf)> ...
 188:  7*  [residue-alt-confs 4 "B" -10000 #f]
In /home/paule/autobuild/Linux-kalypso-pre-release-gtk2/share/coot/scheme/coot-utils.scm:
1456:  8   [reverse ...
1457:  9*   [f ...
1457: 10*    [residue-info 4 "B" -10000 #f]
/home/paule/autobuild/Linux-kalypso-pre-release-gtk2/share/coot/scheme/coot-utils.scm:1457:21: In procedure SWIG_Guile_scm2newstr in expression (residue-info imol chain-id ...):
/home/paule/autobuild/Linux-kalypso-pre-release-gtk2/share/coot/scheme/coot-utils.scm:1457:21: Wrong type argument in position 1: #f

----
Tue Jun  9 06:30:16 BST 2009

   In a MET, move the SD to 2.85A from the CG/CE and then do rotamers.
   The SD and CE do not move. [HK]


----
Tue Jun  9 06:30:23 BST 2009

   MolProbity display clashes with Alt conf atoms handled correctly,
   currently it's a mess. [HK]


---
Tue Jun  9 06:30:49 BST 2009

   Update symmetry on Ctrl-G resno Return [HK]


----

     Think about providing a syncing "service"

     sync from here:
 
     rsync is basically a free lib?

     rsyncing from an ftp server directory tree?


----

    Charles want to add "update mechanisms" to the "testing" talks
    day.


    Display the R-factors from the refmac log file somehow.
    Job for Bernhard?


----

   Phil: gross structural changes in 7A anisotropic data (had higher
   resolution structure in a different conformation) for Den.


----

    When cootaneer fails, what has it done?  Why did it fails?


---

    Think about "undo" on delete molecule form the display manager

    validation -> check waters or density fit window, there are
    multiple copies of these that fall behind the main window

   Using GNOME suse 11?

-----

   Nick: compare pdb file to original sequence.  As a validation tool

* Done.

----

   Easy way to make the baton build build backwards.

* Done.

---

  Phil (considering building from scratch ) "Join up these
  fragments for me"


---

   fobs - fobs difference map made in coot, you need to define a scale
   for the maps.

	minimize:   sigma (fobs1 - (k+b)fobs2)

   * consider phase?  |FWT| can substitue for |Fobs| I guess.
     (considering phase only makes sense for isomorphous maps)

----

   shift-click and ctrl click on the file browser for file selection.

* Done.


----

In tests:

Fix non-printable characters.

Entered testcase - Read small molecule .res file


WARNING:: failed to get AFIX handle for  <DA> HIS  N   ,
WARNING:: failed to get AFIX handle for  <DA> HIS  CA  ,
WARNING:: failed to get AFIX handle for  <DA> HIS  C   ,
WARNING:: failed to get AFIX handle for  <DA> HIS  O   ,
WARNING:: failed to get AFIX handle for  <DA> HIS  CB  ,
WARNING:: failed to get AFIX handle for  <DA> HIS  CG  ,



----

    NCS Jumping with ligands: pdb2vt4.ent (from EDS) [Beta1 Adrenergic
    Receptor]


----

    INFO:: when adding a Gtk2-only widget, you need to adds it
    create_xxx_dialog function to add-c-extern.sh


---
Wed Jul  8 12:44:43 BST 2009

	links should be dashed.  1ehz.

* Done.

----
Wed Jul  8 12:45:53 BST 2009

   sequence view problems 1ehz

* Fixed.

----
Mon Jul  6 20:04:39 BST 2009

    What's the Ubuntu way of checking for dependencies?  (i.e. those
    in the README file)?

* dpkg-query. Done.


----
Wed Jul  8 23:29:46 BST 2009

	 the tick marks for crashes-on-sphere-refine-maybe.pdb are not good.


----
Sun Jul 12 12:07:56 BST 2009

	how does add_alt_conf residue range work?

   It doesn't!  The function is not filled.


----
Mon Jul 13 09:16:18 CDT 2009

   Project for summer student:

   improve b-factor validation test by finding the proper distribution
   of b-factors for residues (and include as a function of resolution)
   (and refinement program for amusement).

  Another:

   Code up Otwinowski and Richardson distributions for use in
   Ramachandran refinements.  Compare and contrast.  Do it well, and
   there could be a paper in this.

  Another:
 
   Add Pre-pro to Rama plot possibilities (may need to enhance
   clipper).


----
Mon Jul 13 11:42:00 CDT 2009

   molecule_class_info_t's get_residue() function has arguments in the wrong order.


----
Tue Jul 14 15:11:53 CDT 2009

  distances should make reference to the molecule from which them came
  and should go away when molecule is undisplayed or deleted.


---
Tue Jul 14 19:19:07 CDT 2009

	in backrubbing 
float 
molecule_class_info_t::backrub_rotamer(const std::string &...
  
    can we (should we) delete the fragment_asc?


---
Thu Jul 16 09:21:32 CDT 2009

   what is 'errat' (validation of protein structures)?

   what is 'modloop' - modelling of loops in protein structures.

---
Thu Jul 16 13:57:19 CDT 2009

   The ligand fitting dialog in windows (i.e. the dialog of fitting
   ligand positions after the ligand fitting has completed) is
   missing.

---
Sun Aug  2 21:52:04 BST 2009

   Use scm_from_locale_string() not scm_makfrom0str().


----
Mon Aug  3 12:29:37 BST 2009

   Show (4 sigma by default) outliers in the geometry analysis. 

----
Wed Aug  5 18:28:07 BST 2009

   side-chain weighting (slider?), useful for low-res refinement.

---
Tue Aug  4 13:04:58 BST 2009

   INFO: install coot (autobuilder) dependencies on Ubuntu 9.04: -> 76
   packages.


---
Sat Aug  8 09:10:22 BST 2009

openexr fink on mac uses OpenGL and frameworks.  


---
Mon Aug 10 14:18:16 BST 2009

   Why does Ca + Ligands show the SAH in 1o9s, but not the SAM in in
   1nw3? [Steve Bellon]


---
Tue Aug  4 13:04:58 BST 2009

   side-chain weighting (slider?), useful for low-res refinement.


---
Tue Aug 18 08:42:28 BST 2009

   When adding waters, do a sphericity test.  You need to know the
   resolution of the data and have pre-calculated tables of sphericity
   at various resolutions.  So that requires a search of PDB data.

---
Thu 20 Aug 2009 10:13:21 EDT

   JB et al: 

   Copy fragment 2bz1 //B/1-305  -> no atoms selected 

   Make the image type png

   Don't display bumps - in the environment distances (so that
   biologist see the hydrogen bonds - but not the bumps (the are not
   interested in them)).

   Unphosphorylate (change the residue type)

   Add a key binding to cut side chains (shift K, perhaps).

* Done.

   Perhaps the hydrogens on a ligand from probe is not working as
   you'd hoped.  Try with several different ligands.  See "Use
   write_connectivity to generate ligand hydrogens" in the rel-todo.

   Then you will need to use the hydrogen conn file when running
   reduce.

   It is reasonble for me to say he period is 0 and the sd is 0 on a
   rotatable bond.  So coot should deal with that.   

      i.e. if the sd is 0, then no torsion restraint on that tor.
       also if the period is 0, there souold ben no torsion
       restraints.

   The weight matrix refinement control dialog needs to be read/set.

   We want button in the main toolbar to open a coordinaes file (so
   that we dont have to click and hold and select a menu item on the
   file men (Word does it).  Same for cif and mtz.  We need to make
   the open_coords_dialog() acceccesable from scripting.

   split residue (his) and rotamers (try both from CA, and all). 

   Try rotamers on a alt confed residue.

   Delete a water (shift W).

   On adding a water, it defaults to adding to "Waters molecule" or
   something - stop doing that and if it is default/not assigned, then
   give me a dialog to assign it

   Somehow go to reside "GTY" (let me type a triple letter).

   The sequence alignement seems to fail, because lets say the protein
   starts with residue 11, the sequence alignemnt should be for the
   model, not the sequence.  Needs investigating.

   Residues with missing atoms: I want to use the . and , key on the
   generated list.

   Same goes for the list from Find Waters.  I want a list of waters
   in a dialog that I can use . and , with. Like questionable waters.

   Backrub rotamer search on a PHE in a helix.  -> terrible solution.
   Need to include clash terms. 

   User-define-click resets the cursor after it has run the
   function. This can be inconvenient, when user-defined-click calls
   itself at the end (e.g. we don't know how many times
   user-defined-click needs to be called - so we use it in one-by-one
   mode)

   Try sphere refinement with hydrogens on. Check that convergence
   happens.

   "SSM all" to this: give me a dialog, with an "all on" button.  

   chains ligands
   A        D
   B        E
   C        F

   SSM-all should copy across the ligands too - that is what we are
   most interested in. You will have to find which ligands should come
   with which chains.

   NCS ghosts don't work with ligands/cofacors.  Can you make it work
   with those too?

   There should be quick way to undo/undisplay "highight intereting
   residue"

   Make an uberfile with views, maps, proteinm models in it.

   align vs pir sequence failed to do anything obvious.
   We then dis alignment-mismatches-gui by hand and it gave us a gui.

   An "Insert residue" should take you to the nearest residue.
  
   But the residue number is wrong for insertions.
   Mutations look fine.

   In the colour map rotation for maps, let there be "an allowed range"
   on the colour wheel).

----

      In baton build, make it work better/more clearly when starting
      with no skeleton.
      ->  ask me to make a skeleton

* Done.

     Doing go to blob when the refiement map has not been set, 
     needs better feedback.

* Done.

     Sharpening a map, read in from a map:  -> crash.

* Fixed

     I want a function to set the refinment weight for me, 
     i.e.  it should do a triple refine and look at the chi squareds.
     Keep doing that with different weights until we get to 1.0.

* Scheduled.

     The very low level map from phenix, makes refinement fail to 
     converge without setting the map weight to 3000!

     How about doing Ramachandran refinement with optional sphere refinement.

---
Sun Aug 23 16:43:37 EDT 2009

    make_asc(CMMDBManager *mol) should be (also?) a constructor of an asc.

----
Mon Aug 24 16:57:58 EDT 2009

    When doing add strand here, make the residues ALA

----
Mon Aug 24 16:58:02 EDT 2009

   quit with no save state.

----
Mon Aug 24 17:02:29 EDT 2009

   3 molecules loaded,  0 3 and 7

   post-go-to-atom-window starts up showing 0.

   Use the option menu to change to 3.

   -> fails to update the residue list

   Use the option menu to change to 0.
   Use the option menu to change to 3.
	
   -> updates the residue list.

    Probably due to this: 

    set-go-to-atom-molecule 3 (before post-go-to-atom-window)
 
    but go to atom dialog starts up showing molecule 0.

 
----
Mon Aug 24 17:17:35 EDT 2009
  
    how do I start up coot and force it to read the state script? 

---
Mon Aug 24 17:17:40 EDT 2009

   can you make the display manager start up the "correct" size?
   (is it saved in the state script?)

----
Mon Aug 24 17:24:50 EDT 2009

   Difference map peaks with a nice ligand, completely connected at 5
   sigma, there are 10 "peaks" on the ligand! when we peak search
   difference map.

----

   simple average of maps, many maps, something like difference-map function.

* Done.

---

   set map is *not* difference map!  need this function

---
Mon Aug 24 17:42:55 EDT 2009

   "library of things in crystallization: what is this blob: let me
   trivially select."

   cryo-solvent molecule dialog:  Where is this?


----
Mon Aug 24 17:52:10 EDT 2009

   Close the application with the X, don't save the state. 

   Perhaps, needs thinking about.


---
Mon Aug 24 17:53:32 EDT 2009

   On exit (or optional exit) save all modified moelcules with default
   save names.
  

---
Tue Aug 25 10:09:07 EDT 2009

  check the Chi-2 on a leucine.  The dictionary is CA-CB-CG-CG2, but
  built-in is...?  And how does that work with molprobity rotamers?

  These also may be suspicious:
  THR VAL LEU ASP ASN GLU GLN ILE


---
Tue Aug 25 10:54:57 EDT 2009

    Pukka puckers?  Dont use the out of plane distance, use the N1/9
    C1' vector and find the distance of the Phosphate from the
    extension of that line.  It is more sensitive than the out of
    plane distance.

    > 3.0A -> C3' says Jeff.


 Change the probe radius you can see potential interactions at longer distnaces.
 Jeremy: Try changing this at quarter angstrom increments.

 David: how about looking at a "packing moment" - perhaps having to
 move a block first.  Looking athe the dots along the edge of range of
 residues, moving the range of reisdues by a rotation/trans operator
 derives from the dots.


----
Tue Aug 25 14:30:19 EDT 2009

    2b4j: do a sequence association and add terminal residue (either
    at end or a test deleted residue)


----
Tue Aug 25 14:30:19 EDT 2009


    set-hardware-stereo-angel-factor : typo somewhere?  preferences?

* Done. c-interface.cc

----
parse-pisa-xml.scm :371 should check for #f in status


148  in coot-utils.scm   

0 "E" -10000 #f 

wrong type in position 1: #f 

----
Wed Aug 26 05:00:42 EDT 2009

   Align & Mutate changes the residue numbers 
      
   Align & Mutate seems to do a rotamer fit of all residues, I didn't ask for that
 
   In Align & Mutate, if I do not click the "autofit residues" nothing
      seems to happen, the sequence is not mutated.  Oh, or possibly
      it mutated residues in the wrong place.


----
Wed Aug 26 10:28:45 EDT 2009

 with spw:

  Allow me to have more cycles before saying "Time's Up"

  Allow me to interupt a bad refinement with "Esc" key.


---
Wed Aug 26 17:12:43 EDT 2009

   get_rotamer_name_scm() needs to be passed an alt_conf?


----
Wed Aug 26 17:18:58 EDT 2009

   fix_nomenclature_errors() works on residues with alt confs?

* Punt.

   rotamer_graphs() for residues with altconfs.

* Seems OK.

   molecule-info-class's set_residue_to_rotamer_number() works with altconfs?

* No.  The alt conf is not specified in the c-interface.h function (even).


----
Sat Aug 29 06:47:18 BST 2009

   get rid of GTK_ENABLE_BROKEN


---
Sun Aug 30 10:21:50 BST 2009

   JED request:
   extend the output capabilities of export-map so that it covers a
   molecule (say).  "auto-extend".


----
Tue Sep  1 19:17:15 BST 2009

   Put phosphate backbone mode into symmetry.

* Done.

----
Fri Sep  4 13:40:49 BST 2009

   It would be nice if the alignment results (e.g Validate vs PIR or
   Align & Mutate) came up in a slider


---
Fri Sep  4 16:18:20 BST 2009

	Project student idea: In refining RNA, it would be nice to
	have extra restraints to keep the base planes aligned.  This
	could be done with a combination of angle and plane (pseudo)
	restraints as a mmCIF link.  Write a program to generate these
	from a PDB file.  You'll need a wiggle-room factor.  Maybe
	generate an intermediate representation/file that the user can
	optionally edit and generate the dictionary from that.

Garib: 08/01/09


  If top/par file could be converted to the following type of
  instructions then you do not need to define everything in cif file
  (these are for torsion angles, all other restraints can be defined
  similarly)

  General torsion angle restraints for any quartet of atoms:

  external torsion first chain [ch] residue [res] insertion [ins] atom
  [n] [altecode [a]] next chain [ch] residue [res] insertion [ins]
  atom [n] [altecode [a] ] [symm y/n] next chain [ch] residue [res]
  insertion [ins] atom [n] [altecode [a] ] next chain [ch] residue
  [res] insertion [ins] atom [n] [altecode [a] ]
 [symm y/n] value <v> sigma <s> period> <p>

  Exampl

  external torsion first chain A residue 220 atom C next chain A
  residue 220 atom CA next chain A residue 220 atom C next chain A
  residue 221 atom N value -60 sigma 10 period 1



----
Fri Sep  4 16:59:29 BST 2009

   It would be nice if coot::util::model_sequence() (or its argument -
   vseq) only allowed residues that were protein amino acids - thus we
   would not get problems of trying to mutate a SO4 into an ALA in
   Aligment vs PIR.  (Yes, it is a stupid idea to put phosphates into
   protein chain, but some people do it).


---
Sat Sep  5 10:17:51 BST 2009

   in molecule-class-info-mutate.cc, align_and_mutate()

   bool renumber_residues_flag = 0; // make this derived from the GUI one day


---
Tue Sep  8 21:14:49 BST 2009

    Summer student project: Use PDB to find distribution of B-factors
    around centre residue atoms.  Find outliers.  Make validation
    tool.


---
Tue Sep 15 11:55:50 BST 2009

   It would be nice to add the Ramachandran angles to the residue info
   dialog and a colour patch (like refinement dialog).


----
Sat Sep 19 18:11:45 BST 2009

	box-pack-end the "Play Views" button to the vbox (it's
	returned), rather than add it as a normal button (and
	currently new views go after the "Play Views" button which is
	ugly.


----
Sat Sep 19 18:36:06 BST 2009
	
    The bond parameters and bond colours dialogs need to be consolidated.

----
Mon Sep 21 15:43:47 BST 2009

     student project:

     In jiggle-fitting ligands, mask the density by the surrounding
     atoms of the molecule that it will be merged into.  (and/or use
     them to make a bump/clash score).

     Go through the ligands of the PDB, how well does this method work?


----
Tue Sep 22 17:24:48 BST 2009

     http://www.chembiogrid.org/index.html 
     What services do they provide that we can use?  SMILES->3D?


---
Wed 23 Sep 2009 22:08:41 BST

   check the fps of rnase coords, say, when it is molecule 0 and when
   it is molecule 10, with the others closed or undisplayed.


----
Thu 24 Sep 2009 18:07:02 BST
   
   Student project:
   The use of sound as a form of user-feedback in molecular
   model-building

---
Mon Sep 28 14:26:25 BST 2009

   molecule_class_info_t::find_ncs_matrix(int SelHandle1, int SelHandle2) const;
   does not pass the graphics_info_t::ncs_matrix_flag.  It should (and
   should not use it statically).


---
Tue Sep 29 11:39:05 BST 2009

   Doing a rotamer search should first to an automatic nomenclature
   check (and correction if necessary), e.g. 105A in 1a0j.

---
Tue Sep 29 13:18:33 BST 2009

   move minimol into coot-utils and make that just one coot-utils lib.


----
Wed Sep 30 14:55:21 BST 2009

    Several tool bar function (autofit rotamer, rotamers, edit chi
    angles, pepflip, side-chain-180, edit-backbone,
    mutate&auto-fit-rotamer, simple-mutate, add-peptide, add-alt-conf)
    might work better using active-atom.

    Big change.


----
Wed Oct  7 15:44:59 BST 2009
   
   Ingo K: if you try to add a c-terminal residue, coot does recognize
   the residue number correctly, but will not do any real space
   refinement. if you output the coordinates, you will see that the
   new residue is attached to the end of the B chain, i.e. BEHIND the
   waters.  i would guess that this breaks real space refinement in
   coot. i would guess ... (it asks to increase
   max_number_refine_residues or so)

----
Sat Oct 10 08:36:45 BST 2009

   Student project.  Add metal distances (water validation).
   consolidate water validation.  Run difference map water validation
   across all structures in the PDB with anomalous data. See "metal
   distances" above.


----
Mon Oct 12 19:07:07 BST 2009

   Get real-space R factors (or correlations) per residue for many
   structures in the PDB (as a function of resolution say).  Use these
   to validate *this* structure.


---
Sat Oct 17 05:43:53 EDT 2009

    on starting up fresh coot:

(set-default-initial-contour-level-for-difference-map -1075912448.00)
(set-idle-function-rotate-angle -1075912448.00)

set-default-bond-thickness needs to be added to the history


----
Wed Oct 14 11:05:03 BST 2009
 

   student project: implement hole:

    O.S. Smart, J.M. Goodfellow and B.A. Wallace (1993) 
    The Pore Dimensions of Gramicidin A 
    Biophysical Journal 65:2455-2460. 

http://d2o.bioch.ox.ac.uk:38080/doc/index.html
http://d2o.bioch.ox.ac.uk:38080/


----

    student project:

  o nucleic acid analog of Ramachandran plot (theta, eta), see message
    from Bill Scott:
    http://www.csb.yale.edu/people/pyle/software/index.html 


----

     student projects:

  o NOE violations (or successes)

----
Mon Oct 19 10:09:42 EDT 2009

   o maybe base-pair alignment (and preparation of pseudo-restraints)
    is a student project.

   o use the sphere tool (interactive position and radius) to select a
     set of residues - then use those for rigid-body refinement.

----
Fri Oct 16 14:38:00 BST 2009

  o INFO:: mac compile problems: 
    check out the static analyser:
    http://arstechnica.com/apple/reviews/2009/08/mac-os-x-10-6.ars/9


----
Mon Oct 19 10:39:58 EDT 2009

    The baton build problem:  Gina

   Produce like this (not verified):  
   Calc -> Map Skeleton -> On

   Ca Baton mode mode, open, close, open, undo, close
   [candidate positions displayed]
   Ca mode, 
   Undo 
   Accept 
   Undo 
   Accept -> Now the baton positions atoms are gone.


----
Tue Oct 20 11:29:07 EDT 2009

   Need to add Pre-PRO to clipper.
 

---

   consider looking for unsatified H-bond donors and acceptors, for
   NMR structures in particular c.f. cing? 


----
Wed Oct 21 13:56:38 EDT 2009

   For VTF: Talk with Kevin about running sequins on the Chang
   structures (backwards and forwards).


----
Wed Oct 21 16:49:43 EDT 2009

   For VTF: make sure that the findligand is usable with absolute
   levels of density.


 
----

    Jane: some comment about someone who made a server? program that
    searches for (general?) active sites in proteins.
   
    Find out more.


---
Sat Oct 24 00:30:39 EDT 2009

    Filling density with close dummy atoms and then refining them: 

    Acta Cryst. (1997). D53, 540-543  
    Local Improvement of Electron-Density Maps
    A. G. Urzhumtsev

    Improves weak loop density?


----
Sun Oct 25 17:54:59 EDT 2009

	
(from Garib)
>
> libcheck
>
> file_l <dicntionary_file>
> mon <monomoer_you_want>
> nodist y


----

   INFO:: Friesner, R. A et al Glide: a new approach for rapid,
   accurate docking and scoring, 1. Method and assessment of docking
   accuracy J. Med. Chem 2004, 47, 1739-1749

----
Sat Nov 14 19:58:48 GMT 2009

   needs cleanup:

   in several of the delete_x() functions,
   e.g. delete_atom_by_atom_index, delete_atom, delete_residue, we
   have:

   CResidue *residue_p =
      graphics_info_t::molecules[imol].get_residue(resno, ins_code, chain_id);
   if (residue_p) {
      coot::residue_spec_t spec(residue_p);
      g.delete_residue_from_geometry_graphs(imol, spec);
   }

   This needs to be rationalised into (perhaps) 
   delete_residue_from_geometry_graphs_maybe(chain_id, resno, ins_code);



----
Sat Nov 14 12:56:21 GMT 2009

   Take a shelx molecule and add 10 to the coords (means fixed).  Now
   display symmetry.  Fail.  See message from Z Dauter.

----
Mon Nov 30 10:23:48 GMT 2009

   check that users of the fill_chain_option_menu function work
   nicely.  They now should use menu_item_label(item) in the callback,
   not the horrible casting of pointer to user_data to char *s.


---
Tue Nov 24 17:07:20 GMT 2009

    An extension to backrub rotamer: 

    For each rotamer: try chi-by-chi to improve the fit to the density
    of the side chain.  It will probably only need 2 rounds.  Needs
    non-trivial minimiser though (having said that, it's only a series
    of 1-d functions).

    Hmmm... what to do about clash_score...?  The variation in clash
    score wil be relatively low and ignorable?


---
Mon Dec  7 16:32:38 GMT 2009

Ed Pozharski:

   Coot severely slows down the system when I try to generate the
   density fit graph using multi-model (NMR-style) pdb file.  First,
   it apparently generates as many dialogs as there are models, yet
   every graph has multiple bars (which is not very useful since they
   are overlapping).  Computer becomes unresponsive for some time, but
   when it comes back everything is very slow (except that I can
   rotate and zoom in smoothly, things get slow if I re-centre).

   The problem appears to be that coot eats a lot of RAM under these
   circumstances.  I have 4Gb here and it's still not enough so
   machine starts using the swap and thus gets very-very slow.
   Interestingly, most of the extra RAM is owned by Xorg, not coot,
   and is not freed even when I exit (but it is when I log out).

   If I load the same number of models as individual pdb-files, I can
   generate multiple density fit graphs with virtually undetectable
   increase in memory use.


---
Mon Dec  7 17:39:09 GMT 2009

   do the 1 2 3 4 keys work for edit chi angles?


----
Mon Dec  7 17:43:08 GMT 2009

   rotate peptide to optimise a combination of the density fit of
   mainchain (3) atoms and the ramachandran plot.

----
Mon Dec 21 11:14:22 GMT 2009

    In the Go To Atom, rightmouse should activate a pop-up, 

    Highlight "MET"s  (for example) taking the residue type of the residue label we clicked on
    Highlight type -> entry

---
Tue Dec 22 08:03:46 GMT 2009

   o (set-find-hydrogen-torsion 1) and edit chi angles on a SMILES
     CCCCCC.  The torsions are in the wrong order.  ::: 4


---
Sun Dec 27 02:09:22 GMT 2009

   o the arguments of molecule_info_class_t get_residue() are the
     wrong way round.

* Done.


---
Mon Dec 28 02:23:49 GMT 2009

   Problem?

   Try refining a post-reduced model (e.g. tutorial-modern.pdb)
   reidues 1A, what is happening to the hydrogens?

---
Tue Dec 29 14:44:49 GMT 2009

   Improve the NBC restraints so that regularization doesn't cause
   clashes.  eg. regularize A 45-47 with
   (set-do-probe-dots-post-refine 1)


---
Fri Jan 15 21:06:12 GMT 2010

   double-clicking on go to atom residue, elg. E2B from 3hb5, should
   take you to the atom closest the the centre of the residue, not the
   first residue (it doesn't have a CA of course).


----
Sat Jan 16 09:16:32 GMT 2010

   tutorial example 

   1bav, Difference map peaks, Bad rotamers, missing NCS waters


---
Sun Jan 17 12:51:25 GMT 2010

   Simple mutate base in 1v15.  This has PDB standard base names?


----
Thu Jan 21 11:31:07 GMT 2010

   o Read in coot-ccp4/monomer-BIP.pdb and regularize.  -> No restraints.
     Dynamic add does not use libcheck.  should it?
     ::: 1


----
Thu Jan 21 23:23:30 GMT 2010

   o Refine: click on ASN, click on terminal NAG, get residues that
     are linked and use "sphere" refinement for just those residues.
   

---
Mon Jan 25 15:10:19 GMT 2010

	on finding a glycosidic linkage, Coot should do the DEL-xxx
	mod if needed.

---
Mon Jan 25 17:22:45 GMT 2010

	sphere refine on a water (with no neighbours). -> No
	restraints dialog.  [JED]

---
Tue Jan 26 16:47:45 GMT 2010

    When doing refinement, check that all the non-H atoms of the
    selected residues appear in the restraints.  If not, give a dialog
    and run a hook (which is passed a list of the failing residues
    specs and types). [JED].


---
Wed Jan 27 01:47:31 GMT 2010

    Make sure guile is 1.8.7 and GSL is 1.13


---
Sat Feb  6 16:58:24 GMT 2010

   Small baton build tutorial:

   Build the kinase consensus sequence peptide: 2q0n


---
Sat Feb 13 14:35:12 GMT 2010

   INFO:: 3CH6 is nice for showing NCS skipping.  Or is it a case of
   the wrong spacegroup? NCS jumping.

   Look at the ligand near 177


---
Mon Feb 15 02:32:00 GMT 2010

   (Refmac) Additional restraints:

exte dist first chain A resi 2 atom CA second chain A resi 5 atom CA value 4.0 sigma 0.02 
http://www.ccp4.ac.uk/dist/html/refmac5/keywords/keywords_5_5.html#External

----
Wed Feb 17 03:05:21 GMT 2010

   Sphere refinement with alt confs creates problems (no progress).
   e.g.: around B 425 in 3h7w.

----
Fri Feb 26 11:21:07 GMT 2010

  
   All pairs NCS rmsd comparison. (Requested by FvD)

  GK:

   - my favourite plot would be the "CD plot" (but then again, it
   would, wouldn't it?) - see for instance:

   http://xray.bmc.uu.se/cgi-bin/gerard/image_page.pl?image=usf/pics/cdplot_1ldn.gif


----
Sat Feb 27 13:01:00 GMT 2010

   Missing waters using this PDB file:

   tutorial-modern-symmetry-2-bad-water-finding-using-this-protein.pdb


---
Wed Mar  3 14:16:25 GMT 2010

  INFO:: To stop being in DOS mode:

  M-x set-buffer-file-coding-system 
  undecided-unix

  -> save


----
Thu Mar 25 15:55:10 GMT 2010

   EJD: let me click on one atom in a chain to convert it (the whole
     chain) to mainchain.


---

   here's one for the validation tutorial: 1m1f


---
Thu May 13 13:36:31 BST 2010


   JED comments:


   - Ball and stick does not obey show hydrogens off.

  Agreed. This should be fixed.

   - Environment distances are shown for hydrogens -- makes the view
     pretty crowded.

  Don't know how to "fix"

   - The "Apply" button in the bond parameters window seems superfluous.

  Agreed, it's a bit non-canonical.

* Done.

----
Mon May 17 16:46:04 BST 2010

  Andrew Sharff:

    CMH
    BEZ and BOX
    many: methyl mercury

* a problem for PDBe, not Eugene.         

----

   Andrew Sharf::

   Just refining / building a structure that has a number of LINK and
   SSBOND cards in the PDB header.  I have had to renumber one of the
   chains to make the numbering consistent with other structures.

   However, on saving output, the residue numbers in the SSBOND and
   LINK lines is not changed.


----
Tue May 18 11:25:15 BST 2010

----
   BN: When alt-confing ligands, I'd like to see a slide-bar to set
       the occs.

----

   On loading the output of refmac, give us a way of seeing the "most
   moved" atoms. (seems to be a useful thing to see when developing
   lbg).

----
Fri May  7 12:48:16 UYT 2010

    re: Kate's project: c.f. SPASM (USF?)

----
Thu May 20 18:38:52 BST 2010
  
   o Bob Nolte:  "clean up structure button"

	 reorder chains - renumber waters - arrange water around
	 protein - fix nomenclature errors - merge water chains

----
Sun May 23 13:35:38 BST 2010


   2WOT: lbg the ligand and re-import back to Coot main.  Now try to
   fit the ligand.

   It's hard.  Make it easier. Identify matched atom (atom name map) and 
   the rotatable bonds and match them between reference ligand and new 
   ligand.  Now set the rotatable bonds to those values (before 
   minimisation).

   o need some cleverness to match and set the chi angles of the 
     rotatable bonds (before lsqing, even, I guess)  (or at least, 
     an LSQ is needed afterwards).


---
Thu May 20 18:45:29 BST 2010

    put in Edit Peferences the settings for these:

    (define *clear-out-backup-run-n-days* 7)
    (define *clear-out-backup-old-days* 7)


---
Thu May 20 18:50:50 BST 2010

    o Bob Nolte: give me a slider to set the alt confs.

---
Thu May 20 18:58:26 BST 2010

   o I want be able to edit bond colours and params from the molecule
     in the display manager.


----
Sun May 23 13:37:55 BST 2010

   Ian Tickle:

   o I have long been advocating the use of the RMS density
     deviation Z-score statistic for this purpose, i.e.
     sqrt(sum(delta_rho2)/Npoints)/sigma(rho), where delta_rho is a
     grid-point value in a 2(mFo-DFc) map, the sum is over the region
     of interest (atom, residue, side-chain etc), Npoints is the
     number of grid points in the region, and sigma(rho) is the
     uncertainty in the density (which must be estimated for the
     asymmetric unit of the map, NOT just the region).  Note that it's
     equivalent to the negative log-likelihood, which IMO is the
     appropriate statistic for this situation, if it's assumed that
     the errors in the density have a normal distribution.


-----

    integration of calcuated chemical shift with molecular dynamics... 
    interesting..

----
Fri Jun  4 13:24:37 CEST 2010


Judit Debreczeni wrote:

> 1. coot remembers the presence of additional representations from the
> state file. It would be nice if it also remembered if they were
> switched on or off.

Funny - I was annoyed by that lack of feature just last night.

* Done now.

> 2. If there are two different ligands bound to the same active site
> with partial occupancies (with alt conf identifiers A and B), coot
> will still draw the environment distances between them, making the
> view quite crowded.

Bleugh.  quite right. 


> 3. For covalent ligands (but also for SS-bonds etc): if I change the
> chain ID, the pdb header will still refer to the original chain IDs in
> therms of LINK and SS-bond records.
> It would be nice if the header could be automatically updated. (Or if
> not, it would be nice to have a "view and edit pdb header" option
> somewhere.)

Second person, maybe 3rd including Eleanor, to complain about this.
Agreed.


> 4. Currently it is frustratingly tedious to set the occupancy for
> newly added alt conf ligands. 

Do you know about:

(set-add-alt-conf-new-atoms-occupancy 0.2)

before splitting?

If that won't do, a trivial function to fix occs of given alt conf can
be made by wrapping set-atom-attributes


> It would be nice if we had a similar
> mechanism to what is used for standard residues (without the rotamer
> bit, of course) -- i.e. a slide bar or some such.

Yes.  That can be done - even with scripting - by above system.

-----
Wed Jun  9 10:40:31 CEST 2010

   Topf et al structure (2008?)

   Flex-EM, hinge-based flexable fitting of protein structures to EM
   map (decent/good resolution maps).

----
Wed Jun  9 12:13:44 CEST 2010

   Eddy Arnold

   "strategic flexibility" (wiggling and jiggling)

   Janssen et al. 2005, J. Med Chem. 48, 1901 (perhaps)

   hydrophobic pocket - adaptabilty to changes:
    
   2004 Das, 

   2008

   ligprep in schrodinger suite, preparing for glide

   make the "omega" geneation of ligand conformers usable at the
   scripting layer

    ,... e.g give me 2000 low energy conformers.

	Janssen:  Centre of Molecular Design


----
Wed Jun  9 13:17:27 CEST 2010

   Open Display manager.
   Add an add rep (ball & stick, say).
   Turn off all add reps of this molecule.
   Close display manager.
   Open Display Manager

       -> the "All" button has disappeared.
   

----
Fri Jun 11 11:56:15 CEST 2010


   debora makino - elena conti's lab.


------

> How do you shutdown coot such that it doesn't test whether backups are
> needed?
>
> (define clear-backups-maybe (lambda () (delete-coot-backup-files
> 'delete) #f))

Add to docs.

* Done.

----

1T5S nice map to play with?

  TT:
  flood fill density with dummy atom, 
  move the atoms the centre line.
  find characteristic ca-ca-ca-ca geometry.
  link bits together.  15 residues/second

------
Thu Jul 22 21:40:12 EDT 2010

  John Spence: Molecular movies:
   -> make it easy to make movies from a series of electron density 
      maps (mtz files).  

      Consider interpolation.

----

   Update NCS maps based on local NCS. 
   generate NCS matrix by local LSQing.
 
-------

  Garib: new refmac restraints type:
  Kullback - Leibler divergence.

-----
Fri Jul 30 14:54:42 BST 2010

   Fed 12 bins don't run on Fed 13.


----
Thu Aug  5 05:23:49 BST 2010

   Use rdkit/openbabel for Gasteiger-Marsili partial charge calculation


-----
Thu Aug  5 06:32:43 BST 2010

   Do I have the surface normals reversed?
   Try reversing the normals - does that make backface culling work?  

-----
Mon Aug  9 08:12:35 BST 2010

    o CA Zone + ligands has thick hydrogen sticks (e.g. with-ligand.pdb)

* To fix this requires a big change in the colour system of coot.  The
  xth colour is grey and gregy is this (hydrogen sticks) unless
  make_graphical_bonds is called with "no_thinning" (CA mode).  Tricky.


-----
Sun Aug 15 12:59:12 BST 2010

    o in edit chi mode, if I mouse over another chi box then that bond
      gets highlighted.  It shouldn't update hilighted bond when I am
      in edit chi mode.  Or something. See JED mail from a few days ago.


-----
Wed Sep 29 10:35:58 BST 2010

Transparent sphere with glutsphere:  

   martinm:

   You can do it by enabling culling, and using two draw calls - one
   with glFrontFace(GL_CW) and another with glFrontFace(GL_CCW). First
   draw call will draw back faces of sphere, and second one front
   faces, therefore you will get transparent sphere with correct
   polygons rendered from back to front.

http://www.opengl.org/discussion_boards/ubbthreads.php?ubb=showflat&Number=250770


----
Thu Sep  2 11:50:17 BST 2010

  Use GL_ONE_MINUS_DST_COLOR in blending with white background edges?

http://msdn.microsoft.com/en-us/library/ms537046.aspx

----
Thu Sep  2 11:40:03 BST 2010

   Libraries have been installed in:
       /lmb/wear/emsley/coot/fed-12/lib
    
    use the `-Wl,-rpath -Wl,LIBDIR' linker flag


-----
Interesting valgrind problems:

==23673== 67,982 bytes in 2,432 blocks are possibly lost in loss record 9,100 of 9,127
==23673==    at 0x4C28CC1: operator new(unsigned long) (vg_replace_malloc.c:261)
==23673==    by 0xFE8ED98: std::string::_Rep::_S_create(unsigned long, unsigned long, std::allocator<char> const&) (in /usr/lib/libstdc++.so.6.0.13)
==23673==    by 0xFE8F8B4: ??? (in /usr/lib/libstdc++.so.6.0.13)
==23673==    by 0xFE8FA52: std::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string(char const*, std::allocator<char> const&) (in /usr/lib/libstdc++.so.6.0.13)
==23673==    by 0x6437F07: coot::protein_geometry::simple_mon_lib_chem_comp(CMMCIFLoop*) (protein-geometry.cc:1128)
==23673==    by 0x643D3FA: coot::protein_geometry::init_refmac_mon_lib(std::string, int) (protein-geometry.cc:281)
==23673==    by 0x643F0F9: coot::protein_geometry::init_standard() (protein-geometry.cc:2751)
==23673==    by 0x4A5E2C: graphics_info_t::init() (graphics-info.h:856)
==23673==    by 0x4A257B: main (main.cc:269)
==23673== 

another:

==23673==    by 0xFE8F8B4: ??? (in /usr/lib/libstdc++.so.6.0.13)
==23673==    by 0xFE8FA52: std::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string(char const*, std::allocator<char
> const&) (in /usr/lib/libstdc++.so.6.0.13)
==23673==    by 0x6437EDB: coot::protein_geometry::simple_mon_lib_chem_comp(CMMCIFLoop*) (protein-geometry.cc:1124)
==23673==    by 0x643D3FA: coot::protein_geometry::init_refmac_mon_lib(std::string, int) (protein-geometry.cc:281)
==23673==    by 0x643F0F9: coot::protein_geometry::init_standard() (protein-geometry.cc:2751)
==23673==    by 0x4A5E2C: graphics_info_t::init() (graphics-info.h:856)
==23673==    by 0x4A257B: main (main.cc:269)
==23673== 

another:

==23673== 86,435 bytes in 2,432 blocks are possibly lost in loss record 9,106 of 9,127
==23673==    at 0x4C28CC1: operator new(unsigned long) (vg_replace_malloc.c:261)
==23673==    by 0xFE8ED98: std::string::_Rep::_S_create(unsigned long, unsigned long, std::allocator<char> const&) (in /usr/lib/libstdc++.so.6.0.13)
==23673==    by 0xFE909E9: std::string::_M_mutate(unsigned long, unsigned long, unsigned long) (in /usr/lib/libstdc++.so.6.0.13)
==23673==    by 0xFE90B7B: std::string::_M_replace_safe(unsigned long, unsigned long, char const*, unsigned long) (in /usr/lib/libstdc++.so.6.0.13)
==23673==    by 0x6437FFD: coot::protein_geometry::simple_mon_lib_chem_comp(CMMCIFLoop*) (basic_string.h:970)
==23673==    by 0x643D3FA: coot::protein_geometry::init_refmac_mon_lib(std::string, int) (protein-geometry.cc:281)
==23673==    by 0x643F0F9: coot::protein_geometry::init_standard() (protein-geometry.cc:2751)
==23673==    by 0x4A5E2C: graphics_info_t::init() (graphics-info.h:856)
==23673==    by 0x4A257B: main (main.cc:269)
==23673== 

Another:

==23673== 2,502,720 bytes in 2 blocks are possibly lost in loss record 9,123 of 9,127
==23673==    at 0x4C28939: operator new[](unsigned long) (vg_replace_malloc.c:305)
==23673==    by 0x552A776: CIsoSurface<float>::returnTriangles(clipper::Xmap<float> const&, clipper::Coord_frac const&, float, coot::Cartesian) const (CIsoSurface.cpp:1576)
==23673==    by 0x552B3D6: CIsoSurface<float>::GenerateSurface_from_Xmap(clipper::Xmap<float> const&, float, float, coot::Cartesian, int) (CIsoSurface.cpp:616)
==23673==    by 0x6ED8A1: molecule_class_info_t::update_map_triangles(float, coot::Cartesian) (molecule-class-info-maps.cc:534)
==23673==    by 0x6EDC91: molecule_class_info_t::update_map_internal() (molecule-class-info-maps.cc:256)
==23673==    by 0x6EDCAE: molecule_class_info_t::update_map() (molecule-class-info-maps.cc:244)
==23673==    by 0x5D74A4: graphics_info_t::update_things_on_move() (graphics-info.cc:2095)
==23673==    by 0x5D74E8: graphics_info_t::update_things_on_move_and_redraw() (graphics-info.cc:2086)
==23673==    by 0x4F4ADC: set_rotation_centre (c-interface.cc:2726)
==23673==    by 0x59B473: _wrap_set_rotation_centre(scm_unused_struct*, scm_unused_struct*, scm_unused_struct*) (coot_wrap_guile_gtk2.cc:11925)
==23673==    by 0x87E633D: deval (in /home/paule/autobuild/Linux-scylla-pre-release-gtk2/lib/libguile.so.17.3.1)
==23673==    by 0x87E732A: scm_i_eval_x (in /home/paule/autobuild/Linux-scylla-pre-release-gtk2/lib/libguile.so.17.3.1)

----
Sun Oct 10 12:14:21 EDT 2010

   o trivial but useful addition, but a button "non-standard residue"
     at the bottom of the residue list when mutating.
     ::: 0.5

-------
Sat Nov 13 16:34:37 GMT 2010

     o JED: environment distances when there are hydrogens on the ligand should not 
       show interactions with the hydrogen base atom when there is an interaction
       with the hydrogen.

     o JED: Display manager: all molecules to xxx representation, (option menu, say).

     o JED: load 2wot: then sprout hydrogens on the ligand -> total failure

     o JED: keybinding w should check that there is a water/pointer-atom molecule defined.

     o JED: Sphere refine on a water should do something sensible.

     o (There should be a key that goes to hetatms)
 
     o question to self: do we need restraints to layout molecules?  e.g. 2bzh


----
Mon Nov 15 16:27:35 GMT 2010

    A 500 CDK in 2xmy fails to prodrgify (alt confs)


----
Wed Nov 17 14:01:30 GMT 2010

    Bernhard's Biocryst report.


    o Out of plane distances should have graphical output c.f. a distance.

    o 'findligand --help' does not list x-pos, y-pos,... (there may be more) 

    o we may want to lose the HID part of the tutorial


----
Sat Nov 27 10:58:27 GMT 2010

  o merge a 3GP into tutorial modern then try to undisplay hydrogens
    with Bond Parameters.  Fail (because of new bonding algorithm) - not for 0.6.2.


----
Sat Nov 27 10:58:27 GMT 2010

  o ncs difference graph updating on autofit rotamer and refine residue

---
Tue Nov 30 02:42:51 GMT 2010

  o The hydrogens on DTT are drawn wrongly (additional half-bond).  

* Too complicated for now

----
Tue Dec  7 00:37:42 GMT 2010

    student project:

   dig out whatcheck parsing interface, parse more info and
   automatically fix stuff (needs (more) intelligence).


----
Tue Dec  7 15:25:30 GMT 2010

    Here is how to run molprobity probe for just the interactions of
    environment onto the ligand (and vica versa):

    (No internal interaction of ligand)

   Here residue 459 is the ligand:

   probe -u -both "459" "not 459" 2gpu-A459.pdb 2gpu-coot-0.pdb > probe.dots

   Do this after running reduce of course (so that the Hs are pointing
   in good directions).

---
Wed Dec  8 14:37:58 GMT 2010

  Francis Reyes
   
  On pressing Esc during refinement: 

	Can we have a dialog box pop up summarizing the geometry and
	asking us to accept/reject instead of reverting back? 
	Sometimes a real space refine takes too long/gets lost when
	the best fit to the density is observed before that happens.


---
Mon Feb 28 17:01:48 GMT 2011

   in make a difference map (say for overlayed ligand structures)
   find the scale the maps (around a given spot) mean, standard
   deviation perhaps? 


----
Sun Jan  9 17:21:10 GMT 2011
 
   How about this: map the surface of a protein onto a cylinder, 

   Then unwrap the cylinder and display as a 2d plot.

   Do that for both sides of an interface.

   or mutant vs native.

   so we will need some way to mark the surface of atom selection.



----
Mon Jan 10 10:50:29 GMT 2011

    after changing the interface, e.g. c-interface.h

    you will probably need to do:
    an install in scheme
    an install in python
    run swig to make new coot.py
    an install in src, to update installed coot.py


 
---
Sun Mar 20 23:08:50 GMT 2011

checkmol/matchmol

GPL Pascal

http://merian.pch.univie.ac.at/~nhaider/cheminf/cmmm.html


---
Sun Jun 19 14:15:19 BST 2011

    JED:

    Add an tool to add OP3 to DNA and RNA. Find the torsion of the OP1
    and OP2. Average.  Add 180 then position OP3 by torsion,angle,bond
    (use the values in the dictionary).

    (ideal DNA and RNA helix builder doesn't add them - you might like
    to reconsider that).

---
Sun Jun 19 14:18:41 BST 2011
 
    JED:

    Tool to WC base-pair a residue range and put the results into the
    same molecule (as the reference) but a different chain of course.

* Done.


----
Wed Jul 20 19:34:25 BST 2011

    Ctrl Shift middle-mouse behaves differently to Shift Ctrl middle-mouse


----
Fri Jul 22 16:43:47 BST 2011

    Make a small tutorial for jligand link making.  

    Find something from the PDB?

    Check jligand tutorial first?


---
Sun Jul 10 23:29:32 BST 2011

   Note: 
   When making a scrolled list to add things in to, don't forget to put
   the vbox into a viewport and the viewport into the scrolled window
   (e.g. nomenclature errors dialog).

---
Mon Jul 11 11:58:13 BST 2011

   2y0z has problematic Z40 on reading.  Why?


----
Wed Jul 20 13:03:27 BST 2011

   ca+ ligands view mode should not be able to click on waters (or
   things that are not ca/ligands!)


----
Tue Jul 26 23:11:17 BST 2011

    Mutate and autofit to an MSE should change the atoms to HETATM,
    but it does not.

    So, minimol's atom class needs 'bool is_hetatm' flag.  

    Import: minimol's residue constructors from CResidue * need to set
    the is_hetatm flag by testing at->Het.

    Extraction: When constructing the pcmmdbmanager(), we need to set
    at->Het = true when making the CAtom *.


---
Thu Jul 28 14:26:13 BST 2011

    Need to import and display the refmac external restraints (from a
    file)


----
Thu Jul 28 14:42:51 BST 2011
  
    Must have user defined colour system! (for reading prosmart
    output)
	
    here: green (good alignment), red (bad alignment), white (not aligned)


----
Fri Jul 29 19:48:03 BST 2011

    Importing an sdf/mdl file should create a residue with HETATMs!

* Done, I think - needs checking.

    Get the TLC from the first line (if the first line consists of 3 letters)


----
Fri Aug 12 15:59:29 BST 2011

   Areej bug: Adding an altconf on a ligand (hetgroup) makes the
   atoms into ATOMs, (they were HETATMs)

    or is it that ligands with alt confs are drawn without dictionary?

----
Fri Aug 12 16:05:07 BST 2011


   Nice feature:

   In the sequence view (the window needs a title) it would be nice if
   the ligands had a highlighting background colour and showed the TLC
   (and molecule name?) as a tooltip.
    
   Yes.  Very nice.

----
Fri Aug 12 16:07:05 BST 2011

   Areej Bug. (Another one).

   Add a Glycerol using Solvent Ligands.

   Add another.  

   They go into different chains.

   Badness


----
Fri Aug 12 16:17:52 BST 2011

    In the Solvent Ligands dialog, DTT should be added.

    And it would be nice if the molecule names were there too.

---
Fri Aug 12 12:29:03 BST 2011

   INFO:: For new depositions:

   Parse this:
   http://www.ebi.ac.uk/pdbe/searchResults.html?display=latest&tab=pdb
   http://www.ebi.ac.uk/pdbe/searchResults.html?display=latest&tab=pdb

   Or this, perhaps better: The rss feed:
   http://www.ebi.ac.uk/pdbe/pdblatest/latest_pdb.xml

   Then get:
   http://www.ebi.ac.uk/pdbe-srv/view/files/$code.ent
   http://www.ebi.ac.uk/pdbe-srv/view/files/r{$code}sf.ent

   Then cif_to_mtz 

* Done!  For the most part...

---
Mon Sep  5 16:19:45 BST 2011


dbi for COD:


guile> (use-modules (dbi dbi))
guile> (define db-obj (dbi-open "mysql" "cod_reader::molecules:tcp:www.crystallography.net::")
... )
guile> db-obj 
#<guile-dbi open mysql cod_reader::molecules:tcp:www.crystallography.net:: (0 . db connected)>
guile> db-obj
#<guile-dbi open mysql cod_reader::molecules:tcp:www.crystallography.net:: (0 . query ok, got results)>
guile> (display db-obj)
#<guile-dbi open mysql cod_reader::molecules:tcp:www.crystallography.net:: (0 . query ok, got results)>
guile> 
guile> (dbi-query db-obj "select * from numbers")
guile> db-obj

(write (dbi-get_row db-obj)) (newline)
(write (dbi-get_row db-obj)) (newline)
(write (dbi-get_row db-obj)) (newline)
(write (dbi-get_row db-obj)) (newline)

----
Wed Aug 17 16:05:36 CEST 2011

Some ideas for Control key bindings:

    C-c to copy residue 

    Shift C-c to copy a molecule

    C-v to paste residue/molecule to new molecule

    C-x to delete residue
   
    Shift C-x to delete (undisplay) molecule.


----

  learnable (discoverable),
  usable (get stuff done with least amount of clicking).
  forgiving (undo), 
  repsonsive (feedback), 
  asthetics (pretty)


---
Thu Aug 18 16:25:07 CEST 2011


   3pcq for FEL data (photosystem I)


---

   c.f.  Pioglitazone with Rosiglitazone


----

   Discuss with Max: 3s5x and its refmac LINKR records

----

   3oeg: the BCLs at the end of the chain: can we navigate to them
   correctly in density fit and geometry graphs?

----

   "At least one person should look at the map" - GWK

   1qw9

----

   Dale Tronrud structures: 1qlw - nice difference map peaks, waters,
                                   phosphate, disulphide

                            3ls9 - lots of model-building errors, 
                                   missing ligand? badly built loop
                                  THR-ILE mutation

                            3oeg - BCL fantasy.


                            2qns - retracted structure - lots of substantial 
                                   mistakes good for tutorial?


----
Wed Oct 26 13:49:32 EDT 2011

	predict3d for membrane proteins?


----
Fri Oct 28 11:20:15 EDT 2011

   ligand expo -> browse -> popular drugs


----
Sat Oct 29 16:36:25 EDT 2011
  
   allow colour by c-a deviation , c-b deviations etc.  (user-defined scale and 
   colour map of course)

-----
Thu Dec  8 17:09:55 GMT 2011

    interesting glyco:  3u2s, 1s4p


    2fbj fails - goes into neighbour density.


----
Sat Dec 31 18:09:09 GMT 2011


    Failing ligands:

	4a55 A 2062. P08 Prodrgified.  Ring oxygen has 3 bonds.
[19:27:05] Explicit valence for atom # 10 O, 3, is greater than permitted


 PLP  in 3zzj 1246 A


---
Sun Jan  8 12:07:43 GMT 2012

   to get the lbg icon labels on Ubuntu:

   gconf-editor desktop -> gnome -> interface -> toolbar_style "both"


----
Tue Jan 10 06:03:39 GMT 2012

   The new way to set data, set user data on a gtk object:

   gtk_object_set_data(GTK_OBJECT(dialog), "goograph", this);

   // and in the static callback function:
   coot::goograph *g_p = (coot::goograph *) gtk_object_get_data(GTK_OBJECT(widget), "goograph");


---
Fri Jan 13 01:17:45 GMT 2012

    For GLSL
    Triangle order optimization:
    http://gameangst.com/?p=9
    http://www.lighthouse3d.com/tutorials/glsl-core-tutorial/vertex-shader/


---
Fri Jan 13 01:23:15 GMT 2012

   Conversation with Bernie:

   Is it possible to track screen X,Y of the mouse when over chi angle
   buttons and selectively only change to a new selected (highlighted)
   chi angle if delta(Y) > delta(X)?

   Something for 0.8?

* Done, I think.  Or at least, tinkered with...


---
Tue Jan 24 19:15:37 GMT 2012

gentoo libtool usage:
 * Running libtoolize --copy --force --install --automake ...


---
Mon Jan 30 11:45:55 GMT 2012

   Mogul chemistry perception

      determined using input file (mol, mol2, pdb) information.

      http://www.ccdc.cam.ac.uk/support/documentation/mogul/mogul/mogul.3.12.html

      There is no guarantee that bond-type deduction or hydrogen-atom
      addition will be completely correct. Consequently, you should
      check the changes that Mogul has made. A summary of the changes
      made will be given in a Structure edited pop-up.


----
Fri Feb  3 16:36:37 GMT 2012

   relative rpath: 

  http://groups.google.com/group/gnu.gcc.help/browse_thread/thread/1b713af28905247b?pli=1

  man ld.so

http://support.bibblelabs.com/webboard/viewtopic.php?p=64073&sid=04162de5073a8e07db92d0a0c22e23b6

contains($$list($$[QT_VERSION]), 4.2.*) {
    QMAKE_LFLAGS+=$${QMAKE_RPATH}\\\$\$ORIGIN/../lib
} else {
    QMAKE_LFLAGS+=$${QMAKE_RPATH}\\$\$ORIGIN/../lib
}

http://enchildfone.wordpress.com/2010/03/23/a-description-of-rpath-origin-ld_library_path-and-portable-linux-binaries/

consider chrpath

http://stackoverflow.com/questions/6323603/ld-using-rpath-origin-inside-a-shared-library-recursive

    how about this:
   LIBS=-Wl,-rpath,'$${ORIGIN}/../lib -Wl,-rpath,/usr/lib' 


---
Tue Jan 31 22:44:31 GMT 2012

rdkit

python $RDBASE/rdkit/Chem/Features/ShowFeats.py --writeFeats --fdef=$RDBASE/Data/BaseFeatures.fdef 1ezq_RPR.sdf



---
Fri Feb  3 11:46:48 GMT 2012

on windows:

ECHO %HOMEDRIVE%%HOMEPATH%


----
Tue Feb 14 11:54:31 EST 2012

   need a test that updating restraints from mogul works with atom names that are not 4-chars.

   mogul::get_bond_type()

----
Mon Feb 20 17:42:28 GMT 2012

   3t3g has a covalently linked ligand?


----

   we have sets of pdb files and cif files generated by different
   methods given the same SMILES string.

   give me a tool that consolidates the names in the files - and
   restraints of course so that they are "the same".



---
Mon Mar 19 16:55:16 GMT 2012

   Interesting project?  Maybe rama plot can be rendered over this?

   http://giv.sourceforge.net/gtk-image-viewer/gtkimageviewer_tutorial.html

---
Wed Apr  4 15:19:44 BST 2012

  ---> we drop the image url

  such as 

  http://www.chemspider.com/ImagesHandler.ashx?id=24747352&w=200&h=200

  extract chemspider_id = 24747352

  --> we curl get mdl mol file:
  "http://www.chemspider.com/FilesHandler.ashx?type=str&striph=yes&id=" + chemspider_id


  and while you are at it, you might get the wikipedia page, look for 

  href="http://www.chemspider.com/Chemical-Structure.24747352" to extract the chemspider_id


---
Tue Apr  3 17:04:40 BST 2012

valgrind:

 valgrind --leak-check=full --suppressions=../../coot/trunk/guile.supp --suppressions=../../coot/trunk/valgrind-python.supp --error-limit=no --log-file=valgrind.log .libs/lt-coot-real 

# valgrind --leak-check=full --suppressions=../../coot/trunk/guile.supp --error-limit=no --log-file=valgrind.log .libs/lt-coot-bin 
 valgrind --leak-check=full --suppressions=../../coot/git/coot/guile.supp --error-limit=no --log-file=valgrind.log .libs/lt-coot-bin 


  Where are the python suppressions?

   http://stackoverflow.com/questions/16659781/memory-leaks-in-gtk-hello-world-program

   For debugging glib/gtk programs I would use this command:

   G_SLICE=debug-blocks valgrind --tool=memcheck --leak-check=full <gtk program>
   G_SLICE=debug-blocks will turn off gtk's advanced memory management to allow valgrind to show correct results.
   --leak-check=full will show stack traces for the leaked memory blocks.

   You can also use --show-reachable=yes to see stack traces for all
   memory blocks that haven't been free when the program exits.

   There is also the massif valgrind tool that tracks memory usage to
   show which parts of the program are using the most memory.

   Run program under massif:

   G_SLICE=always-malloc valgrind --tool=massif --detailed-freq=2 --max-snapshots=400 --num-callers=20 <gtk program>

   Show results:

   ms_print massif.out.<pid>
     
----
Mon Apr  9 16:45:32 BST 2012

   COOT_CHEMICAL_FEATURES_DEF file needs to be put in place (or the
   env var to be defined in coot script).  See sdf-interface.cc:get_feature_factory().

---
Tue May  1 12:29:50 BST 2012

   To get and set string on widgets use
   g_object_set_data()
   and 
   g_object_get_data()

   coot::dictionary_residue_restraints_t *t =
      (coot::dictionary_residue_restraints_t *) g_object_get_data (G_OBJECT (chooser), "restraints");


----
Tue May  1 15:37:30 BST 2012

  RDKit UFF atom types? UFFGetMoleculeForceField()


---
Thu May  3 18:45:01 BST 2012

    energy-lib.cc

	 // FIXME, allow bond definitions with no distances?
	 if (0)
	    std::cout << "WARNING reject energy lib bond \"" << atom_type_1
		      << "\" \"" << atom_type_2 << "\" \"" << type << "\"" << std::endl;



----
Fri May 18 12:34:38 BST 2012

// I think that graphics_info_t::Geom_p() should be passed to this function.
// 
void
molecule_class_info_t::makebonds() {


----
Tue May 29 19:35:49 BST 2012

    projects: 

    make graphics look nice on white background

    Add data analysis graphs sigF vs F etc.

    Learn Qt.  Make ramachandran widget - with atom selections

    Transparent surfaces via z-sorting


----
Fri May 18 12:34:38 BST 2012

// I think that graphics_info_t::Geom_p() should be passed to this function.
// 
void
molecule_class_info_t::makebonds() {


----
Sun May 20 22:34:54 BST 2012

can redefine functions be done with a macro? using symbol->string and string->symbol?


----
Tue May 22 06:34:20 BST 2012
	
   Burrow-owl:

   Needed starparse, g-wrap-1.9.14, guile-cairo-1.9.91, guile-gnome-platform-2.16.2   
   and guile-1.8.8.  Becuase guile does not have the same prefix as
   target, we need to set PKG_CONFIG_PATH.  guile-cairo is needed to
   get gtk parts of guile-gnome-platform to compile and install.

   guile-gnome-platform needs indent.

prfx=$HOME/burrow-owl
export PATH=$prfx/bin:$PATH
export STARTPARSE_LIBS=$prfx/lib
export STARTPARSE_CFLAGS=$prfx/include
export PKG_CONFIG_PATH=$prfx/lib/pkgconfig:/usr/lib64/pkgconfig
./configure --prefix=$prfx

----
Tue May 22 11:07:38 BST 2012

	fast lennard jones, fast double sqrt.

----
Thu May 24 10:58:11 BST 2012

   get rid of "Won't read preferences" output.

---
Sat May 26 10:36:23 BST 2012

   Newcastle: add a slide or 2 on restraints: planes and torsions
   (different from molecular mechanics - c.f. antechamber).
   43 non-elemental atom types in GAFF, 93 in refmac


----
Fri May 25 14:46:38 BST 2012

	alt conf ligands 3ljj

* Done.


----
Sun May 27 22:55:54 BST 2012

    Here's an interesting function:

    given a molecule (model) (and an optional atom selection?) implement
    get_distances(min_dist, max_dist, max_n_distances) 
 

    which finds all atom-atom distances in the atom selection and
    makes a random selection of them and returns max_n_distances

* Done.

----
Wed Jun  6 13:56:17 CEST 2012

   when generic objects are solid, the point size seems fixed.

----
Fri Jun  8 15:53:56 CEST 2012

QM treatment of large systems

Hisao, Sanchez-garcia, Doerr J. Phys. Chem B 2011 review

consider London dispersion

Goerigk, Grimme 2011, 13, 6670, Phys. Chem. Chem. Phys.


---
Sat Jun  9 17:15:40 CEST 2012

  PDB_REDO add to extension menu next to pdbe getting.

---
Wed Jun 27 10:29:54 BST 2012

   INFO:: format in scheme/guile

(format #t "~3,'0d~%" 5)
-> 
005 

(format #t "~4f%" 6.321567)
-> 
6.322


http://www.gnu.org/software/guile/manual/html_node/Formatted-Output.html#Formatted-Output


---
Wed Jun 27 22:48:01 BST 2012

   building carbohydrate

   nicotinic acetylcholine receptor alpha bungarotoxin: 2qc1 

   a structure with interesting carbohydrate to build:  3zr6
   also 4fom (low res)


---
Wed Jun 27 10:29:54 BST 2012

   INFO:: example of remove and remove_if

#include <vector>
#include <string>
#include <algorithm>
#include <iostream>
#include <functional>

// for erase_if()
class eraser {
public:
   std::string s;
   // the constructor, can be information that needs to be used
   // internally in the operator() function.  This is run once
   // 
   eraser(const std::string &s_in) {
      s = s_in;
   }
   // The passed object type is the <type> of the vector.
   // Run for each an item in the vector to see if it is deletable.
   bool operator()(const std::string &x) const {
      std::cout << "comparing " << s << " vs " << x << std::endl;
      return (x == s);
   }
};

int main(int argc, char **argv) {

   std::vector<std::string> v;
   v.push_back("One");
   v.push_back("Two");
   v.push_back("Three");
   v.push_back("Four");
   v.push_back("Five");

   std::string bad_boy = "Three";

   // remove() and remove_if() generate a useful pointer used by erase()
   v.erase(std::remove(v.begin(), v.end(), "Two"), v.end());
   v.erase(std::remove_if(v.begin(), v.end(), eraser(bad_boy)), v.end());
   
   for (unsigned int i=0; i<v.size(); i++)
      std::cout << "   " << i << " " << v[i] << std::endl;
   return 0;
}

   // also see erase/remove_if usage in morph_fit_uncrunch().


---
Tue Jul  3 17:14:08 BST 2012

CentOS 6 extras

glib2-devel zlib-devel  (coot config)
gtk2-devel

(for coot clipper config)

libXt-devel (for gtkglext)
libgnomecanvas-devel for coot configure 

for mosflm (ccp4)
ncurses-devel

for xemacs: 
yum install Xaw3d-devel libXp-devel xorg-x11-xbitmaps libXaw-devel

bison for rdkit.

---
Wed Jun 27 22:48:01 BST 2012

   a structure with interesting carbohydrate to build:  3zr6


---
Thu Jul 26 19:47:52 EDT 2012

    INFO:: other projects

    Mac Coot, 
    Native Mac Coot
    All-molecule scoring modifications, 2011 presentation follow-up
    cif-merge
    cif-grep
    cif2mtz (clipper)
    understand what's in COD
    look at pi-pi stacking in ligand protein complexes. 
        get statistics for plane distances and angle offsets, 
        so that I can model that distribution to generate a penalty function 
        in the regularizer
    zernike-like moments for ligand fitting

    b value distribution histogram


    --comp_id x 


----
Sun Jul 29 16:31:40 EDT 2012

    Roberto:  

    I want a button that splits a water into A and B conformers 50 50
    occs, 0.7A appart and jiggle-fitted
    
    Call it WASP.


----
Tue Jul 31 08:56:41 EDT 2012

    Contact Alex for LH protein?


----
Tue Jul 31 11:16:57 EDT 2012

   Howard: 
   given an mtz file with anomalous data (deltas) and a pdb file, show
   the anomalous map


----
Tue Jul 31 11:17:55 EDT 2012

   Why is refmac not run after libcheck SMILES?

   4th spears lecture 

   William asbury


---
Tue Aug  7 00:13:03 BST 2012

make coot-user-manual.html
texi2html ../../coot-read-only/trunk/doc/coot-user-manual.texi
*** Duplicate node found: Sectionless functions (in ../../coot-read-only/trunk/doc/cc-interface.hh.texi l. 411)
** Node following `Drag and Drop Functions' in menu `Sectionless functions' and in sectionning `Sectionless functions' differ 
** `Sectionless functions' is up for `ligand-search-make-conformers-scm', but has no menu entry for this node
splitstr='-split_chapter' ; \
	texi2html --version ; \
	if [ $? = 0 ] ; then splitstr='-split=chapter' ; fi ; \
	texi2html $splitstr ../../coot-read-only/trunk/doc/coot-user-manual.texi ; \
	cd ..
1.82
*** Duplicate node found: Sectionless functions (in ../../coot-read-only/trunk/doc/cc-interface.hh.texi l. 411)
** Node following `Drag and Drop Functions' in menu `Sectionless functions' and in sectionning `Sectionless functions' differ 
** `Sectionless functions' is up for `ligand-search-make-conformers-scm', but has no menu entry for this node
make coot-user-manual.html done
dvips -o coot-user-manual.ps coot-user-manual.dvi
This is dvips(k) 5.96.1 Copyright 2007 Radical Eye Software (www.radicaleye.com)
' TeX output 2012.03.09:1647' -> coot-user-manual.ps
</usr/share/texmf/dvips/base/tex.pro></usr/share/texmf/dvips/base/texps.pro>
</usr/share/texmf/dvips/base/special.pro>. 



----
Mon Aug  6 15:00:07 BST 2012

   try to pdbe get 4fts: refmac says:

Fatal error in ccp4spg_register_by_symops

   4e6n fails with label error

---
Thu Aug 30 12:17:31 BST 2012

   what is this? dictionary default values:

    Ponder, J. & Case, D. (2003). Adv. Protein Chem. 66, 27±85.
 
   from the refmac monomer library paper

   It's (from) a book.

----
Sat Sep  1 11:38:51 BST 2012

	add hydrogen bond analysis (understand output of reduce) into
	tutorials (validation?)

----

    things for MRC: 
        carbohydrate building paper
        glsl
        build on a mac
        cif2mtz
        ffearing in low res maps - test
        new structures ligand analysis - moments?
        what's up with simplex refinement?

----
Sat Sep  1 11:58:18 BST 2012

   verify3d for membrane proteins?


----
Thu Sep 13 04:55:20 EDT 2012

  4eg6 is interesting to build - and good example of how to use fixed
  atoms in refinement.


---  

   plif-related

 	1.  Cavasotto, C. "Ligand docking and structure-based virtual
 	screening in drug discovery." Current topics in medicinal
 	chemistry 7.10 (2007): 1006-1014.  Link to SFX for this item

	2.  McInnes, C. "Virtual screening strategies in drug
	discovery." Current opinion in chemical biology 11.5 (2007):
	494-502.  Link to SFX for this item

	3.  Barril, X. "Virtual screening in structure-based drug
	discovery." Mini reviews in medicinal chemistry 4.7 (2004):
	779-791.

----
Tue Sep 18 12:47:57 BST 2012

simplify_numbering() needs a scripting interface - and to work over
all chains.  

   That will give us a function to remove insertion codes


----
Tue Sep 18 20:55:39 BST 2012

  How about a ring flip function where you identify an atom in a ring,
  it checks that there are less than 3 connections to the ring, then
  rotates 180 about one of the connecting bonds (the one with the
  largest number of atoms in the fragment) and back-rotates the
  other torsion (if exists), choosing and torsion that minimizes the
  movement of the atoms.


----
Fri Oct 26 14:07:13 EDT 2012

Dave:
   rotamers buttons should have name, not chi1!

* it does have the rotamer name (now)

Jeff
   tell me about the geometry of this residue, bond angle deviantions
   etc (as a table sorted on Z?)

   building a loop: I have a number of resdues, I want to see the
   geometric distrtions of the residues (c.f. how king does it with
   red springs for bonds, and fan for angles?)

For presentation next time:

   Add "it takes a brain to validate" slide	
   figure for bond and angle restraints, 
   Dig out the bayes slide, mention validation
     introduce probability and likelihood 
   
Does zero-occupancy-residue-range work?


---
Tue Nov 13 15:46:06 GMT 2012

   2xkm is problematic when reading/converting the data.


---
Tue Nov 20 10:27:38 GMT 2012

	
   Phil Evans:

mutate then refine

INFO:: backup file coot-backup/_Users_PhilStuff_Projects_Varp_InmaVR_VR32_SeMet2_s13.pdb_Mon_Nov_19_13:45:35_2012_modification_123.pdb.gz
INFO:: replace_coords: 24 atoms updated.
ERROR:: Null atom_colour_map_p with COLOUR_BY_CHAIN_C_ONLY mode
ERROR:: Null atom_colour_map_p with COLOUR_BY_CHAIN_C_ONLY mode
ERROR:: Null atom_colour_map_p with COLOUR_BY_CHAIN_C_ONLY mode
ERROR:: Null atom_colour_map_p with COLOUR_BY_CHAIN_C_ONLY mode
ERROR:: Null atom_colour_map_p with COLOUR_BY_CHAIN_C_ONLY mode
ERROR:: Null atom_colour_map_p with COLOUR_BY_CHAIN_C_ONLY mode
INFO:: graphics_info_t::clear_up_moving_atoms...


----
Thu Feb 21 12:34:18 GMT 2013

OXT on multi-residue torsion:  OXT is left behind on rotation

----
Wed Apr 17 16:47:04 UYT 2013


   o sc: lawerence and colman jmb 1993

   might be worth a look.


----
Thu May 16 13:29:43 BST 2013

   Find areas of a map that should have model but do not:
   
   o make a partial model, run refmac, make a map, low res filter it,
     mask out the protein parts that we have, fill big blobs with dummy atoms - zero occ
     (these should be added to the partial model for sending back to
     refmac).  Make dummy atoms 1.5A apart.

   -> resulting map should have better density for the missing domain.  Does it?


-----
Thu Jun 27 15:40:01 BST 2013

    from chimera:

    create a new map by placing target points in the map (position
    them like go to blob) then mask a map around these atoms.

    Hmm... this is basically how Water+ works.  No point in making
    another interface.


------
Sat Oct  5 09:29:10 BST 2013

   Problems with python's pip:
tiresias ~/compile/pip-1.4.1> pip install tornado
Traceback (most recent call last):
  File "/home/paule/autobuild/build-coot+rdkit-pre-release-gtk2-python/bin/pip", line 9, in <module>
    load_entry_point('pip==1.4.1', 'console_scripts', 'pip')()
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 357, in load_entry_point
    """Yield named metadata resource as list of non-blank non-comment lines
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 2394, in load_entry_point
    
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 2108, in load
    parse_map = classmethod(parse_map)
  File "/home/paule/autobuild/build-coot+rdkit-pre-release-gtk2-python/lib/python2.7/site-packages/pip-1.4.1-py2.7.egg/pip/__init__.py", line 10, in <module>
    from pip.util import get_installed_distributions, get_prog
  File "/home/paule/autobuild/build-coot+rdkit-pre-release-gtk2-python/lib/python2.7/site-packages/pip-1.4.1-py2.7.egg/pip/util.py", line 17, in <module>
    from pip.vendor.distlib import version
  File "/home/paule/autobuild/build-coot+rdkit-pre-release-gtk2-python/lib/python2.7/site-packages/pip-1.4.1-py2.7.egg/pip/vendor/distlib/version.py", line 13, in <module>
    from .compat import string_types
  File "/home/paule/autobuild/build-coot+rdkit-pre-release-gtk2-python/lib/python2.7/site-packages/pip-1.4.1-py2.7.egg/pip/vendor/distlib/compat.py", line 31, in <module>
    from urllib2 import (Request, urlopen, URLError, HTTPError,
ImportError: cannot import name HTTPSHandler

https://plus.google.com/117010240131763942181/posts/h3AzbmSQU6r

 install openssl-devel, then re-make python


-----
Sat Feb  8 22:57:51 GMT 2014

   Gaussian Legendre Integration:
   http://www.holoborodko.com/pavel/numerical-methods/numerical-integration/

------
Tue Mar  4 14:27:39 GMT 2014

   For ligands at RCSB:

   http://www.rcsb.org/pdb/ligand/ligandsummary.do?hetId=AMP


------
Tue Mar  4 14:27:39 GMT 2014

   The thumbnail for ATP should be:

            O-  O-  O-
            |   |   |
	 -O-P-O-P-O-P=0
            H   H   |
            O   O   O-  

      H = vertical double bond

      It is.

---
Tue Mar 11 08:08:48 GMT 2014

Font stuff:

http://stackoverflow.com/questions/2071621/opengl-live-text-rendering
http://nothings.org/stb/font/

---
Fri May 30 03:55:36 CEST 2014

   mmdb documentation 
   http://www.ebi.ac.uk/pdbe/docs/cldoc/object/cl_obj_surf.html
   http://www.ebi.ac.uk/pdbe/docs/cldoc/object/cl_obj_selfnc.html

   Delete hydrogens:
   (new-molecule-by-atom-selection 0 "//*//*[!H]")

   C-alphas selection
   (new-molecule-by-atom-selection 0 "//*//CA[C]")

---
Tue Jul  8 11:03:57 BST 2014

   glycam: what is this?

---
Tue Jul  8 11:03:33 BST 2014

  Boost for Android?

  http://stackoverflow.com/questions/14036311/official-boost-library-support-for-android-and-ios

---
Sat Jul 19 13:17:01 BST 2014

#include <iomanip> 

   // the width settings specify the *minimum* width (i.e. the field
   must be at least specified number of characters).  If the number is
   longer than the minimum width, it just is wider.

   std::setw() before strings
   std::cout.width() before numbers # this is nonsense I now think. Use setw() for both 
                                      strings and ints

	 // Formatted output:
	 //
	 // For float/double numbers
	 // Using std::fixed the precision is the number of decimal places
	 // use std::right << std::setprecision(3) << std::fixed << f
	 // You may need to set std::left at the beginning of the line (typically for strings)
	 // To reset from std::fixed:
	 // std::cout.setf(std::ios::fixed, std::ios::floatfield);
	 // See dictionary-bond-distributions.cc

   For integers:

   #include <iomanip>
   std::cout << "map " << itm->first <<  "  [" << std::setw(2) << itm->second.size() << "] ";


---
Sat Jul 19 13:31:04

 BST 2014

 ipython:

 http://archive.ipython.org/release/2.1.0/ipython-2.1.0.tar.gz

---
Fri Jul 25 13:56:46 BST 2014

  Store the bond order as an enum, not a string.  That makes
  convert_bond_type() conversion to rdkit type easier.

---
Sun Aug 17 18:10:34 BST 2014

   4mt6 good example for morphing - based no what though? Hmm.

----
Wed Aug 20 00:01:56 BST 2014

   If/when compiling guile-dbi, rename src/guile-dbi.x so that it gets rebuilt.

----
 3b75

 hba1c - or something like that.

----
IUCr Note

   o Sameer: java to javascript compiler - used to make javascript astexviewer

----
IUCr Note: 

   Put the mogul/refmac colour annotations over the lbg representation
   of the figure.  Make it interactive.

----
Tue Aug 12 14:32:32 EDT 2014

   Sameer: Make the FLEV output as coordinates so that we can use it
   to make a figure and make it interactive.

---

   Scoring Protein-Ligand Complex interactions:

      http://pubs.acs.org/doi/abs/10.1021/jm050436v

   Improving the scoring of torsion? KDE by CCDC

      http://pubs.acs.org/doi/abs/10.1021/ci500156d

---


You can obtain the newest version of config.guess and config.sub from the ‘config’ project at http://savannah.gnu.org/. The commands to fetch them are

$ wget -O config.guess 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD'
$ wget -O config.sub 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD'

----

watch this vid:

http://www.youtube.com/watch?v=aMSPwFvlICM

  * it was OK, bit old now.

---


   COD: /lmb/home/flong/COD/cod/cif/1/10/01

---

   How do I write atom names in an sdf file?

---

   o add Water Validation to validation tutorial. What about using the
     anomalous data (Nat Echols is doing this?)

   o Add intro to new validation report

   o 1qw9 bad trp

   o 4q8j many cis-peptides

   o 4d72 real non-pro cis

---

   Read this paper: 

   Acta Cryst. (2007). D63, 42-49  [ doi:10.1107/S0907444906041059 ]

   EMatch: an efficient method for aligning atomic resolution subunits
   into intermediate-resolution cryo-EM maps of large macromolecular
   assemblies

   O. Dror, K. Lasker, R. Nussinov and H. Wolfson

   Synopsis: A method for detecting structural homologs of components
   in an intermediate resolution cryo-EM map and their spatial
   configuration is presented.


---- 

  CSHL 2nd Talk:

  Pirate, Parrot, Buccaneer, Prosmart and refinement(?) 

----

   How can we find compact domains that are misplaced in density?

----
  
   Read this:

   practicalfragments.blogspot.com


----

   SLERPing:

   http://number-none.com/product/Understanding%20Slerp,%20Then%20Not%20Using%20It/
   http://www.cbloom.com/3d/

   Thomas Busser, “PolySlerp, A Fast and Accurate Polynomial Approximation of 
   Slerp”, Game Developer Magazine, February, 2004. 

   http://mollyrocket.com/forums/molly_forum_246.html

   Shoemake, K., Animating rotation with quaternion curves, Computer
     Graphics 19, No 3 (Proc. SIGGRAPH'85), 245-254, 1985.
   Pletinckx, D., Quaternion calculus as a basic tool in computer
     graphics, The Visual Computer 5, 2-13, 1989.


   Smoothing quaternion interpolations:

   http://allenchou.net/2014/04/game-math-interpolating-quaternions-with-circular-blending/

   Do it with Bezier splines?
   http://caig.cs.nctu.edu.tw/course/CA/Lecture/slerp.pdf

----
 
  Refinement of hydrogens:  Allow them to be refined by gemetry gradients
     but not by density map (Garib). 



----
    
   How to do file -> string vector

   std::ifstream f(file_name.c_str());
   if (f) {
      std::vector<std::string> lines;
      std::string line;
      while (std::getline(f, line)) { 
	 lines.push_back(line);
      }
    }


------------

    How to use prefer_python

	 if (graphics_info_t::prefer_python) { 
#if defined USE_PYTHON
	    std::string python_command("xxx()");
	    safe_python_command(python_command);
#endif // USE_PYTHON
	 } else { 
#if defined USE_GUILE
	    std::string scheme_command("(xxx)");
	    safe_scheme_command(scheme_command);
#endif // USE_GUILE
          }

-----

   ligand series:

   Structures have been deposited into the PDB with the following
   codes; 4bf2 (Ask1/Staurosporin), 4bhn (Ask1/Compound5), 4bib
   (Ask1/Compound1), 4bic (Ask1/Compound2), 4bid (Ask1/compound3),
   4bie (Ask1/Compound4).

   http://onlinelibrary.wiley.com/enhanced/doi/10.1002/pro.2298
 

---

   Libs on a Mac:

   https://blogs.oracle.com/dipol/entry/dynamic_libraries_rpath_and_mac


---- 

   librsvg gdk-pixbuf-query-loaders:

   https://www.guiguan.net/fix-python-graph-tool-couldnt-recognise-svg-problem/

later:
Thu 28 Dec 2017 01:32:59 GMT

On Debian 14.04
$ coot
Error loading icon: Couldn't recognize the image file format for file '/home/paule/autobuild/Linux-emsley-vm-ubuntu1404-enhanced-ligand-pre-release-gtk2-python/share/coot/pixmaps/05_Hotel_Icon_Hair_Salon.svg'
Error loading icon: Couldn't recognize the image file format for file '/home/paule/autobuild/Linux-emsley-vm-ubuntu1404-enhanced-ligand-pre-release-gtk2-python/share/coot/pixmaps/3-c-new.svg'
Error loading icon: Couldn't recognize the image file format for file '/home/paule/autobuild/Linux-emsley-vm-ubuntu1404-enhanced-ligand-pre-release-gtk2-python/share/coot/pixmaps/3-c-other.svg'
Error loading icon: Couldn't recognize the image file format for file '/home/paule/autobuild/Linux-emsley-vm-ubuntu1404-enhanced-ligand-pre-release-gtk2-python/share/coot/pixmaps/3-c.svg'
etc
$ apt-get install librsvg2-dev

(maybe it was not the dev package but the libs/common that solved it, librsvg2-common)

----

   using lldb:

   lldb .libs/coot-bin 
   process launch -- --no-graphics

   http://lldb.llvm.org/lldb-gdb.html

    How to run lldb on a core:

   $ lldb .libs/coot-bin -c /cores/core.18439

    How to set a break point in a function:

   break set --name malloc_error_break

    and a member function:

   break set --name coot::restraints_container_t::bonded_flanking_residues_by_residue_vector

----
  
   torsion fit this residue should do a rigid body refinement after each modification, 
   or perhaps store the best torsion set and RBR the best ones.


--- 

Thu Jul 23 12:31:24 BST 2015

   Student/paper project:
   Find low points in map: Look at many mannoses say LSQ plane of
   pyranose, projection of C1-C4 onto the LSQ plane and Plane->O2 for
   "up" - the cross of these will give the other dimension.  Make a
   grid.  Sample map around NAGs, BMA and MANs. Use low points
   (convert to internal coordinates) (making dummy atoms?) to assist 
   score from z_weighted_density_score_new() in multi_residue_torsion_fit_map().
   Low points may be resolution dependent.

---- 

   Student/paper project:

   Where mbf is median b factor 
   What's is the mbf(res)/mbf(protein) for each of the residues in an N-linked tree?
   Is this resolution dependent?


----

    Buckyball coords

    https://gist.github.com/jarvist/4570215

------
 
    Topological RS Chirality

    http://www.chemcomp.com/journal/chiral.htm

    Clever old Labute.

----

   o experiment: use Rdkit with pdb perception to generate molecules and 
     compare them with molecules generated from reading the cif.

------

	Interesting Ligands:

   TF7 in 3nia: can be remodelled as glycerol? (next to real ligand)
                dihydoxy sulfanyl is "strange" chemistry (c.f neighbour ligand)
                not published

------

    Pathology scatter plots: holiday project.  

------

    investigate stl_tree.h

------

    openmp notes: 

    http://www.compunity.org/events/pastevents/parco07/parco_cpp_openmp.pdf

    http://www.viva64.com/en/a/0054/

    http://michaelsuess.net/publications/suess_leopold_common_mistakes_06.pdf

------

    readline:  Add -lcurses to shlib/Makefile:

    SHLIB_XLDFLAGS =  -Wl,-rpath,$(libdir) -Wl,-soname,`basename $@ $(SHLIB_MINOR)` -lcurses
 
    (or change support/shobj-conf which creates this)

    Does this mean I can now use command line editing in coot --no-graphics
    --python?

    e.g. import readline

-----

     to CSHL talk: A front page: accession code, comp-id

-----

     Quickly estimate B-factor of atoms:

     Using PDB data, do a plot of density vs radius for atoms at various 
     resolutions so that 

     rpo(r) -> resolution

     Compare that with what we have in a fragment model.

----

   http://www.opengl-tutorial.org/beginners-tutorials/tutorial-1-opening-a-window/

----

   GLSL cylinders and more:

   http://stackoverflow.com/questions/9595300/cylinder-impostor-in-glsl
 
----

   Modeling Local Structural Rearrangements Using FEP/REST: Application to 
   Relative Binding Affinity Predictions of CDK2 Inhibitors

   http://pubs.acs.org/doi/abs/10.1021/ct300911a?journalCode=jctcce

PDB: 1h1q
PDB: 1h1r
PDB: 1h1s
PDB: 1oiu
PDB: 1oi9
PDB: 1oiy
PDB: 1ioy

   Also: 

   Accurate Binding Free Energy Predictions in Fragment Optimization
   Thomas B. Steinbrecher, Markus Dahlgren, Daniel Cappel, Teng Lin, 
   Lingle Wang, Goran Krilov, Robert Abel, Richard Friesner, and Woody Sherman

   Journal of Chemical Information and Modeling 2015 55 (11), 2411-2420


----

     2bmd resdiue 72 is good for trans peptide restraints testing. 
     kw5124 Touw et al (2015): Trans-cis and peptide plane flips 

----

     sphere and interesting stuff

     http://paulbourke.net/geometry/circlesphere/

----

     How to make a core dump for CCP4 Coot

     which coot
     cp that coot script to local-dir
     edit to remove limit
          hack in value of CCP4_PREFIX
     $ unlimit coredumpsize
     $ gdb /lmb/home/public/xtal/CCP4/ccp4-64/ccp4-7.0/libexec/coot-bin core.14139


---- 

     FLEV and Residue->2D Problems:

     00J in 2xir is a good example for dictionary generation (and slider test?).
     Yes, show this in the LIGANDS meeting (it's good for tutorial too)

     FLEV: 4f6x: ZYL A 301: Mg-O on ligand: make it O- and delete the H.
     R-2D: 3p8x: ZYD A 1:   OH on tail should be HO (also HUD ligand)


---- 

     Dictionary files:

     e.g. ftp://ftp.ebi.ac.uk/pub/databases/msd/pdbechem/files/mmcif/1H2.cif

----

     clustering lib: interesting? https://github.com/dsteinberg/libcluster


----

    FBDD, interesting?

    http://www.slideshare.net/AnthonyCoyne1/fragment-based-drug-discovery

----


   To stop this:

$ coot --no-graphics --python --script comp_ids.py

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "all_comp_ids.py", line 19, in <module>
    m = pyrogen.rdkit_mol_chem_comp_pdbx(cif_file_name_in, comp_id)
TypeError: No Python class registered for C++ class RDKit::ROMol

imort rdkit # we need this of course
# but don't forget:
from rdkit import Chem


-----


when trying to bundle, and run coot-bin:
   
* thread #1: tid = 0x0000, 0x000000010bf485e1 libcoot-bundle.0.dylib`set_model_toolbar_docked_position + 337, stop reason = signal SIGSTOP
use_graphics_interface_state ()
Running python script /Users/pemsley/.coot-preferences/00-console.py
Running python script /Users/pemsley/.coot-preferences/ball_for_sticks.py
Running python script /Users/pemsley/.coot-preferences/coot_preferences.py

** (coot-bin:68769): WARNING **: Widget is NULL: accept_reject_dialog_frame_docked
/Users/pemsley/Projects/build-enhanced-ligand-no-guile-gtk/src/.libs/coot-bin:11: GtkWarning: void gtk_widget_set_sensitive(GtkWidget *, gboolean): assertion `GTK_IS_WIDGET (widget)' failed
  /opt/X11/lib/libX11.6.dylib
                             `
/Users/pemsley/Projects/build-enhanced-ligand-no-guile-gtk/src/.libs/coot-bin:11: GtkWarning: void gtk_widget_hide(GtkWidget *): assertion `GTK_IS_WIDGET (widget)' failed
  /opt/X11/lib/libX11.6.dylib
                             `

** (coot-bin:68769): WARNING **: Widget is NULL: model_fit_refine_toolbar_handlebox
/Users/pemsley/Projects/build-enhanced-ligand-no-guile-gtk/src/.libs/coot-bin:15: GtkWarning: GtkWidget *gtk_widget_get_parent(GtkWidget *): assertion `GTK_IS_WIDGET (widget)' failed
  /Users/pemsley/ccp4/ccp4-src-2015-02-05/ccp4-6.5/lib/libgdk_pixbuf-2.0.0.dylib
                                                                                `I
failed to lookup toolbar

** (coot-bin:68769): WARNING **: Widget is NULL: main_window_model_fit_dialog_frame_left

** (coot-bin:68769): WARNING **: Widget is NULL: main_window_model_fit_dialog_frame

** (coot-bin:68769): WARNING **: Widget is NULL: model_fit_refine_toolbar_handlebox

** (coot-bin:68769): WARNING **: Widget is NULL: vbox1

** (coot-bin:68769): WARNING **: Widget is NULL: model_toolbar

** (coot-bin:68769): WARNING **: Widget is NULL: model_toolbar_vsep_toolitem

** (coot-bin:68769): WARNING **: Widget is NULL: model_toolbar_hsep_toolitem

** (coot-bin:68769): WARNING **: Widget is NULL: model_toolbar_style_toolitem

  * frame #0: 0x000000010bf485e1 libcoot-bundle.0.dylib`set_model_toolbar_docked_position + 337
    frame #1: 0x000000010b8c8673 _coot.so`_wrap_set_model_toolbar_docked_position(_object*, _object*) + 195
    frame #2: 0x00000001062d5901 libpython2.7.dylib`PyEval_EvalFrameEx + 19681
    frame #3: 0x00000001062d07c2 libpython2.7.dylib`PyEval_EvalCodeEx + 1666
    frame #4: 0x00000001062d0136 libpython2.7.dylib`PyEval_EvalCode + 54
    frame #5: 0x00000001062f9544 libpython2.7.dylib`PyRun_FileExFlags + 164
    frame #6: 0x00000001062cc60b libpython2.7.dylib`builtin_execfile + 507
    frame #7: 0x00000001062d5901 libpython2.7.dylib`PyEval_EvalFrameEx + 19681
    frame #8: 0x00000001062d07c2 libpython2.7.dylib`PyEval_EvalCodeEx + 1666
    frame #9: 0x00000001062d0136 libpython2.7.dylib`PyEval_EvalCode + 54
    frame #10: 0x00000001062f9615 libpython2.7.dylib`PyRun_SimpleStringFlags + 149
    frame #11: 0x0000000102bf8591 coot-bin`run_python_script(filename_in=0x00007fc5b29c1150) + 1409 at c-interface.cc:6374
    frame #12: 0x0000000102caaaba coot-bin`try_load_dot_coot_py_and_preferences(home_directory=0x00007fff5d305360) + 5034 at coot-setup-python.cc:187
    frame #13: 0x0000000102ca95d6 coot-bin`setup_python(argc=1, argv=0x00007fff5d306790) + 2230 at coot-setup-python.cc:121
    frame #14: 0x00000001028fd7b9 coot-bin`main(argc=1, argv=0x00007fff5d306790) + 9817 at main.cc:408
    frame #15: 0x00007fff893645c9 libdyld.dylib`start + 1
(lldb) ^D


----
Thu Aug 18 12:03:32 BST 2016

   RNA building test: coot-download/5fw1.ent


---
Wed 24 Aug 2016 00:54:22 BST


Redo the torsion energy/distortion:

http://www.cryst.bbk.ac.uk/PPS2/course/section7/os_cov.html#tor_graph_back


---
Mon Sep  5 23:19:46 BST 2016


http://www.click2drug.org/encyclopedia/chemistry/fda-based-rings.html
https://upload.wikimedia.org/wikipedia/commons/a/a8/Heterocycle_compounds.jpg


---
Thu 24 Nov 2016 16:55:07 GMT

   scheme education:

   http://www.cs.dartmouth.edu/~sergey/cs59/S2016/scheme/continuations.txt

mod SNNmd1 show that I've got the wrong internal structure for plane chem
mods.  That adds and deletes refer to named plane restraints (see the
add and delete atoms).

---
Thu  1 Dec 2016 23:08:27 GMT

   testing for std::thread (needs -pthread?)

   https://github.com/molpopgen/devnotes/wiki/Useful-autoconf-macros

   (not quite what I want, I think)

   Maybe this:

   https://github.com/emcrisostomo/CXX-Autoconf-Macros/blob/master/ax_cxx_have_thread.m4

---
Sat  4 Mar 2017 01:13:08 GMT

   How to set a breakpoint in lldb:

   breakpoint set --method set_dihedral

   puts a breakpoint in all methods/functions called set_dihedral()

----
Sat  4 Mar 2017 12:05:58 GMT

   Can you build it better/faster? recap sequence alignment, "chainsaw" model, HOLE picture

   http://science.sciencemag.org/content/355/6328/eaal4326.full

   The atomic coordinates have been deposited in the Protein Data Bank (PDB) with accession 
   code 5X0M, and the EM map has been deposited in the Electron Microscopy Data Bank 
   with accession code EMD-6698.

----
Thu 23 Mar 2017 12:04:12 GMT

    Gert Vriend contact map exploitation paper, 2017.


----
Tue 18 Apr 2017 16:33:00 BST

    5jik is good/interesting for carbohydrate building - non-oligomannose


----
Sun 28 May 2017 17:09:21 CDT

   Oleg Borbulevych's talk: ONIOM QM/MM DivCon improves clash score
   after their refinement of Astex Diverse set.

   2016 paper

   What would you get (clash score deltas) if you tried coot for hydrogen
   addition and refinement?

----
Sat  1 Jul 2017 17:36:03 BST

   rdkit fragments:

http://rdkit.org/docs_temp/Chem/Fragments.m.html


----
Sat  1 Jul 2017 18:27:09 BST

   Makefile variables $@ $^ $<
   $@ -> target,      $^ -> all dependences,   $< -> first dependency

   https://www.gnu.org/software/make/manual/html_node/Automatic-Variables.html

---
Mon Jun 19 19:46:51 BST 2017

   beta strand: phi=-110, psi=120 from link-restraint.cc

   can you use that with add-terminal-residue-using-phi-psi to make an
   extended polypetide strand?

----
Sun  9 Jul 2017 00:52:16 BST

open drug discovery toolkit - what does it do?

https://iwatobipen.wordpress.com/2017/06/04/open-drug-discovery-toolkit-for-python/

http://www.blopig.com/blog/2016/11/how-to-calculate-plifs-using-rdkit-and-plip/

---
Wed 12 Jul 2017 21:55:28 BST

look this up:

Ho, B. K. & Gruswitz, F. HOLLOW: generating accurate representations
of channel and interior surfaces in molecular structures. BMC
Struct. Biol. 8, 49 (2008).

---
Sun 16 Jul 2017 14:57:51 BST

http://www.boost.org/doc/libs/1_64_0/libs/python/doc/html/tutorial/tutorial/object.html

   Creating boost::python::object from PyObject*

   When you want a boost::python::object to manage a pointer to
   PyObject* pyobj one does:

   boost::python::object o(boost::python::handle<>(pyobj)); In this
   case, the object, manages the pyobj, it won't increase the
   reference count on construction.

    Otherwise, to use a borrowed reference:

    boost::python::object
    boost::python::handle<>(boost::python::borrowed(pyobj)));
    In this case, Py_INCREF is called, so pyobj is not destructed when
    object goes out of scope.

----
Sun 16 Jul 2017 22:43:37 BST

    convert an RDKit molecule to boost::python::object

   	             std::pair<bool, coot::dictionary_residue_restraints_t> rp = 
		     geom.get_monomer_restraints(res_name, imol);
		     RDKit::RWMol rdkm = rdkit_mol(residue_p, rp.second);
                     boost::python::list rdkit_mols_list;
		     RDKit::ROMol *cm_p = new RDKit::ROMol(rdkm);
		     boost::shared_ptr<RDKit::ROMol> xx(cm_p);
		     // maybe I can append(xx) rather than needing this step:
		     boost::python::object obj(xx);
		     rdkit_mols_list.append(obj);

---
Wed 19 Jul 2017 23:41:43 BST

what's this?

   Dixon, S. L. et al.
   PHASE: a new engine for pharmacophore perception, 3D QSAR model development,
   and 3D database screening: 1. Methodology and preliminary results.
   J. Comput. Aided Mol. Des. 20, 647-671,
   doi: 10.1007/s10822-006-9087-6   (2006).

---
Thu Jul 27 19:48:52 BST 2017

   To compile newest Python 2.7.x on Mac, use additional configure argument
   MACOSX_DEPLOYMENT_TARGET=10.10
   to get past
   compile python configure: error cannot compute sizeof size_t

---
Thu Jul 27 20:07:08 BST 2017

   to upgrade ipython using pip:

   $ pip install ipython --upgrade


---
Sat 29 Jul 2017 15:37:06 BST

  read this: https://autotools.io/index.html


---
Fri Aug 11 14:19:34 BST 2017

https://stackoverflow.com/questions/21971449/how-do-i-increase-the-cell-width-of-the-jupyter-ipython-notebook-in-my-browser

~/.ipython/profile_default/static/custom/custom.css (iPython) or ~/.jupyter/custom/custom.css (Jupyter) with content

.container { width:100% !important; }


---
Fri  1 Sep 2017 05:52:26 BST

   server threaded:

   https://stackoverflow.com/questions/15074393/python-multithreading-xmlrpc-server


---
Fri  1 Sep 2017 17:51:38 BST

   WebVR and Coot - I discover:

   https://forums.oculus.com/community/discussion/4232/visualizing-protein-structures

   unitymol, molvr

   http://www.molvr.org/category/downloads/


---
Tue 12 Sep 2017 12:19:45 BST

   Try deleting sidechains of a chain in a cryo em map (I was using chain J from 3771) then
   Extensions -> All Molecule -> Fill partial

   It takes forever (copying the map).

----
Mon Sep 11 13:06:17 BST 2017

Thunderbird says this:

(firefox:11665): Gdk-WARNING **: Native children wider or taller than 65535 pixels are not supported

(firefox:11665): Gdk-WARNING **: Native children wider or taller than 65535 pixels are not supported

----
Fri 29 Sep 2017 06:03:42 BST

 Use -Wl,-rpath (if passing to cc) or -rpath (if passing to ld) as more portable than 
       -R. ephemient Mar 30 '09 at 17:35

 also

 https://www.eyrie.org/~eagle/notes/rpath.html

----
Tue  3 Oct 2017 11:49:21 BST

   Tristan Croll mentions PyOpenCL

   https://mathema.tician.de/software/pyopencl/

----
Wed  4 Oct 2017 06:36:53 BST

   sounds

   https://freesound.org/browse/tags/effect/?page=13#sound


----
Thu 16 Nov 2017 01:48:51 ART

  ligand docking things to read

  https://worldwidescience.org/topicpages/q/qm-polarized+ligand+docking.html

----
Thu 14 Dec 2017 14:45:13 GMT

  o  configuring python with ssl - for pip:  --with-ensurepip=install
     and LDFLAGS="-Wl,--rpath=${AUTOBUILD_INSTALLED}${build_type}/lib" of course

  o apply the patch

  Then copy Modules/Setup.dist Modules/Setup
  Because - it gets overwritten...? Or something?

--- Modules/Setup.dist.orig	2017-12-14 14:24:16.314852316 +0000
+++ Modules/Setup.dist	2017-12-14 14:30:02.601760197 +0000
@@ -215,10 +215,10 @@
 
 # Socket module helper for SSL support; you must comment out the other
 # socket line above, and possibly edit the SSL variable:
-#SSL=/usr/local/ssl
-#_ssl _ssl.c \
-#	-DUSE_SSL -I$(SSL)/include -I$(SSL)/include/openssl \
-#	-L$(SSL)/lib -lssl -lcrypto
+SSL=/usr
+_ssl _ssl.c \
+	-DUSE_SSL -I$(SSL)/include -I$(SSL)/include/openssl \
+	-L$(SSL)/lib64 -lssl -lcrypto


----
Sun 17 Dec 2017 16:17:48 GMT

   o Fix the documentation generation (coot-dox.cfg needs to be generated from coot-dox.cfg.in)
     because the source header files are not in the build directory. Useful?

     http://chris-miceli.blogspot.co.uk/2011/01/integrating-doxygen-with-autotools.html

     ::: 2

----
Sat 30 Dec 2017 20:31:56 GMT

   o Nice to have function: --emdb code which uses files if they are available or gets the
     bundle first if not.

----
Sun 31 Dec 2017 04:54:00 GMT

   o Non-bonded contacts are slow to generate on a big chain (say 400 residues or more) e.g. EMD-6698.

     Make them faster.

---
Sun 31 Dec 2017 05:00:22 GMT

   o put the refinement in a thread.


----
Sun 31 Dec 2017 14:35:22 GMT

File to string;

https://www.fluentcpp.com/2017/12/19/build-strings-from-plain-string-up-to-boost-karma/

  std::ostringstream fileContentsStream;
  fileContentsStream << std::ifstream("MyFile.txt").rdbuf();
  std::string fileContents = fileContentsStream.str();

---
Sun 31 Dec 2017 14:39:12 GMT

  interesting minimization:

http://www.csm.ro/home/mmelsp_series/mmelsp_08_papers/nandrei_08.pdf


---
Sun 31 Dec 2017 20:15:08 GMT

  pango-cairo:

https://developer.gnome.org/pango/stable/pango-Cairo-Rendering.html

  pango notes:

https://www.gtk.org/api/2.6/pango/PangoMarkupFormat.html

---
Thu  4 Jan 2018 16:54:56 GMT

   Have a look at 6aqf

   Allosteric Coupling of Drug Binding and Intracellular Signaling in the A2A Adenosine Receptor

---
Wed 17 Jan 2018 05:07:31 FKST

  jupyter interesting

  import jupyter_client
  k = jupyter_client.KernelManager()
  k.start_kernel()

---
Sun 18 Feb 2018 15:24:11 EET

   HTC Vive on Linux:

   https://www.gamingonlinux.com/articles/first-steps-with-openvr-and-the-vive-on-linux.7229

----
Sun 18 Feb 2018 18:06:29 EET

   For moving single atoms of the moving atoms (for vrcoot), look at
   graphics_info_t::move_single_atom_of_moving_atoms(int screenx, int screeny)

----
Tue 20 Feb 2018 23:47:55 EET

   MDFF example:
 
   http://pubs.rsc.org/en/content/articlehtml/2014/fd/c4fd00005f

   Actually, maybe not - needs close reading.

---
Sun 25 Feb 2018 07:28:36 EET

   cryo-em tutorial:

   o Note: 4m1m  model and 7325 for the map (Gaby)

---
Sun 25 Feb 2018 18:17:33 EET

   o cryo-EM tutorial 6771 middle of J: 196 209 281

     the middle of the pyrimidine ring of the ligand 401 J NDP


---
Mon 26 Feb 2018 10:43:18 EET

   o 3wbm is a good example for RNA building - when RCrane works again :-/
     It has some difference map features that need fixing (in the protein).

---
Fri  9 Mar 2018 14:57:25 GMT

   o Note to self: covalently bonded ligands must be in the same chain as the
     protein chain to which they are attached.


---
Fri  9 Mar 2018 15:03:18 GMT

   o when merging a ligand, try to change the chain-id to that of the protein
     to which it's close. Add checked button to gui: auto-chain-for-ligands


---
Fri  9 Mar 2018 15:12:44 GMT

   o yesterday I installed 3 virtual machines:
     centos 6.9, Fedora live latest, Ubunutu 17.10 (latest)
     All compiled coot out of the box with build-it.
     (I had to install a few system libraries with apt, yum, dnf)


---
Sat 24 Mar 2018 03:00:25 GMT

   o Can you find some "negatively" scoring parts of a domain for jiggle fitting?
     Make n test points inside the domain box and iterate moving them away from the
     closest atom of the protein.  Make a chain of test points so that you can use
     SeekContacts - use a distance that is more than the maximum distance a
     test point will move (n_iter * step size). Atoms that move away from more than
     one atom are interesting (probably). Test points will drift away on the outside.
     So only keep those points that are within xA of any atom.

     Use generic display objects.

---
Sat 24 Mar 2018 10:27:45 GMT

   Stuart: Make a "extensions looker-upper" as a dropdown option in the Coot interface,
           so that people don't have to find/download/run scripts.

---
Fri 20 Apr 2018 14:53:26 BST

   3j3q hiv capsid

   4whj - interesting fitting Alvarez et al, domain needs to be shifted - 4.6A

----
Sun 20 May 2018 12:54:32 BST

   cute graphics: https://github.com/prideout/aobaker
                  https://github.com/prideout/blog-source/blob/master/p61/Cylinder.glsl

                  http://github.prideout.net/
                  http://prideout.net/blog/

    built them into coot :-)



----
Sun 20 May 2018 12:54:28 BST

    OpenGL and gtk on gtkglext mailing list (last entries?)

    https://mail.gnome.org/archives/gtkglext-list/2014-February/msg00016.html


---
Sun 20 May 2018 12:59:28 BST

   sharpconstruct for drawing heads?


---
Sun 20 May 2018 13:22:00 BST

    https://www.blendswap.com for heads

----
Tue 22 May 2018 15:12:04 BST

   o re: a console: look at this:
     http://www.cplusplus.com/reference/cstdio/freopen/
     reopen std out as a static string stream?


	
---
Mon 28 May 2018 12:33:02 BST

    Indirect GLX is no longer supported by default.  You can enable it with:

    defaults write org.macosforge.xquartz.X11 enable_iglx -bool true

   Jeremy Huddleston Sequoia
   https://bugs.freedesktop.org/show_bug.cgi?id=99146


---
Tue 29 May 2018 16:29:39 BST

   investigate __FILE__ __LINE__ and __FUNCTION__ for debugging


---
Fri  1 Jun 2018 21:52:08 BST

    graphics:

    https://cesium.com/blog/2013/04/22/robust-polyline-rendering-with-webgl/

---
Sat  2 Jun 2018 17:26:41 BST


  14.150 How can I make line or polygon antialiasing work?

  https://www.opengl.org/archives/resources/faq/technical/rasterization.htm#rast0150



---
Thu  7 Jun 19:50:40 BST 2018

	compiling on ubuntu 18.04:

   https://stackoverflow.com/questions/39188919/different-exception-specifier-with-g-6-2
   HACK_THROW_

---
Fri  8 Jun 2018 17:31:20 BST

   Keele notes:

   have a look at this:

   3j3q hiv capsid

   4whj - interesting fitting Alvarez et al, domain needs to be shifted - 4.6A


----
Tue 12 Jun 2018 17:35:48 BST

  record the desktop

  ffmpeg -f x11grab -r 30 -i :0.0 /tmp/recording.mp4


----
Thu 21 Jun 2018 14:35:55 BST

   Use PyGI+GTK instead of pygtk with Python3

   Converting gtk_timeout_add:

   https://askubuntu.com/questions/228849/pygtk-to-pygi-gtk-timeout-add-equivalent

----
Fri 22 Jun 2018 14:44:20 BST

  Spin-search Hydrogen atoms - read this first

  Hydrogen bond networks
  R.W.W.Hooft, C.Sander and G.Vriend,
  Positioning hydrogen atoms by optimizing hydrogen bond networks in
  protein structures
  PROTEINS, 26, 363--376 (1996).

---
Fri  6 Jul 2018 11:24:42 BST

    1h4h residue 49 is good for NCS jumping

---
Fri  6 Jul 2018 21:53:36 BST

In refmac


   vdwr 2

divides sigma by 2

---
Sat 14 Jul 2018 02:23:20 BST

   std::unique_ptr: read this:

http://www.drdobbs.com/cpp/c11-uniqueptr/240002708

---
Sun 15 Jul 2018 14:45:10 BST

   converting a QuickTime movie

ffmpeg -i {in-video}.mov -vcodec h264 -acodec aac -strict -2 {out-video}.mp4


---
Wed 18 Jul 2018 02:06:02 BST

 Fitting problem: https://www.nature.com/articles/nature12162

                  https://www.sciencedirect.com/science/article/pii/S0969212608001330


 This?

Electron Cryomicroscopy of Viruses at Near-Atomic Resolutions
Article September 2017
DOI: 10.1146/annurev-virology-101416-041921

Accurate model annotation of a near-atomic resolution cryo-EM map
Article (PDF Available) Proceedings of the National Academy of Sciences 114(12):201621152


 Interesting for other reasons:

The HIV-1 envelope glycoprotein structure: nailing down a moving target
Article
Jan 2017IMMUNOL REV
Andrew B WardIan A Wilson



----

   Henry van bedem?

   Read this:  minor modification in a ligand results in different binding mode.

   j med chem 2016, 59, 6356


----
Sun 22 Jul 2018 13:47:36 EDT

   transparent triangles in raster3d?
   Yes, it is possible.

---

evaluation System and web infrastructure of 3d cryo em model challenge, 2018, JMB

uses EMRinger


----
Mon 23 Jul 2018 13:59:52 EDT

   sgc xchem sabrina mckinnon(?) HAO1 fragment screening.


----
Thu 26 Jul 2018 14:43:36 BST

for difference map for cryoem from refmac:

mapcalc vector


---
Fri 27 Jul 2018 14:45:14 BST

   o Pathwalking: Look at 2015/16 EM model-bulding trial:
     find e2pathwalker_auto.py?

---
Tue 21 Aug 2018 16:20:52 BST

   o CUDA cubic spline for grid interpolation:

     http://www.dannyruijters.nl/cubicinterpolation/

---
Mon  6 Aug 15:21:40 BST 2018


Inverse Kinematics, Featherstone?

   https://github.com/bulletphysics

Also
   Bullet Physics SDK:
   real-time collision detection and multi-physics simulation for VR,
   games, visual effects, robotics, machine learning etc.

Also

   https://dartsim.github.io/ DART (Dynamic Animation and Robotics Toolkit

Book: Rigid Body Dynamics Algorithms 2008, Featherstone, Roy
   A comprehensive collection of the best rigid-body dynamics algorithms

---
Mon 27 Aug 2018 12:21:09 BST

   what is geexlab?

---
Tue 28 Aug 2018 03:27:54 BST

   command line parsing in c++ 11

https://www.reddit.com/r/cpp/comments/4zhm2n/which_library_would_you_recommend_for_parsing/

---
Tue  4 Sep 2018 06:13:12 BST

    read this:

    https://www.ncbi.nlm.nih.gov/pubmed/27318041

    Defining the limits and reliability of rigid-body fitting in cryo-EM maps using
    multi-scale image pyramids.

    and this:

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3150432/

    BCL::EM-Fit: Rigid body fitting of atomic structures into density maps using
    geometric hashing and real space refinement

    read this:

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5715128/

    Tools for the cryo-EM gold rush: going from the cryo-EM map to the atomistic model

    read this

    RosettaES: a sampling strategy enabling automated interpretation of difficult cryo-EM maps.
    Frenz B, Walls AC, Egelman EH, Veesler D, DiMaio F
    Nat Methods. 2017 Aug; 14(8):797-800.

    read this:

    Woetzel N, Lowe EW, Meiler J. Poster: GPU-accelerated rigid body fitting of atomic structures
            into electron density maps. 2011 IEEE 1st International Conference on Computational
            Advances in Bio and Medical Sciences (ICCABS).2011. [Ref list]
            Journal ref is one of the above, I think (BCL::EM-Fit)

---
Tue  4 Sep 2018 16:15:01 BST

   chimera:

   Visualizing density maps with UCSF Chimera.
   Goddard TD, Huang CC, Ferrin TE
   J Struct Biol. 2007 Jan; 157(1):281-7.

   Fitting to maps described here?


---
Fri 21 Sep 2018 10:49:26 BST

    Jane Richardon: packing in 1cgi looks nice.
                    non-Pro cis 3er0 (should be 1 in 3,000 or 1 in 10,000 for non-carbohydrate proteins)
                    cablam outlier: 3ja8 Leu 235

---
Tue 11 Sep 23:10:09 BST 2018


http://opengl.datenwolf.net/gltut/html/index.html

---
Mon Sep 17 21:55:15 BST 2018

   build this from a starting ideal helix:
   (which you might be able to find with a cuda-based search?)
   5toh

---
Mon 24 Sep 2018 15:48:27 BST

   quaternion axis rotation in freeglut 2.8.1 spaceball.c

   void sbrot(int x, int y, int z) {
     float axis_len = (float)sqrt(x * x + y * y + z * z);
     rot = quat_rotate(rot, axis_len * 0.001f, -x / axis_len, -y / axis_len, z / axis_len);

   Does this mean that a spaceball works? Hmm!

---
Wed 26 Sep 2018 22:44:14 BST

 Jeremy Huddleston Sequoia 2016-12-19 19:46:44 UTC

    Indirect GLX is no longer supported by default.  You can enable it with:

       defaults write org.macosforge.xquartz.X11 enable_iglx -bool true

---
Thu 27 Sep 2018 08:12:03 BST

   font stuff

   http://www.lighthouse3d.com/very-simple-libs/vsfl/


---
Thu 27 Sep 2018 11:57:26 BST

   GLSL antialiasing/multisampling

   https://learnopengl.com/Advanced-OpenGL/Anti-Aliasing

---
Thu 27 Sep 2018 15:12:53 BST

   3gp0 needs helix shift of MR solution

   Try low res, prosmart-like restraints or morphing wrapped with prosmart restraints.


---
Thu 27 Sep 2018 15:33:44 BST

   histogram of H (or HA) density values in the apoferritin map.


----
Thu 27 Sep 2018 15:58:58 BST

    Dave Brown suggests:
    ligand validation meeting: what should we do with ligands from pandda - they will
    seem to be rubbish by conventional metrics.

----
Sun 14 Oct 2018 17:10:02 BST

   bfgs

   https://www.caam.rice.edu/~dennis/images/Broyden-Fletcher-Goldfarb-Shanno.jpg

---
Sun  4 Nov 2018 05:04:52 GMT

   atomic compare_exchange_weak
   https://www.codeproject.com/Articles/808305/Understand-std-atomic-compare-exchange-weak-in-Cpl

---
Wed  7 Nov 2018 15:45:36 GMT

   PLIP is a thing (I mentioned it above!)
   https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4489249/
   A 500 in 1osn - what does this give in pli?

---
Sun 11 Nov 2018 16:32:37 GMT

   read this before updating hydrogen bonds:

   Kortemme T, Morozov AV, Baker D. An orientation-dependent hydrogen bonding potential improves
       prediction of specificity and structure for proteins and protein-protein complexes.
       J Mol Biol. 2003;326:1239–1259.

---
Sun 25 Nov 2018 12:26:00 BRST

   porting pygtk to python3  PyGI+GTK PyGObject Introspection
   https://wiki.gnome.org/action/show/Projects/PyGObject/IntrospectionPorting?action=show&redirect=PyGObject%2FIntrospectionPorting


---
Tue 11 Dec 2018 02:47:48 GMT

   differential equation solver

https://github.com/sheljohn/osd

---
Tue 18 Dec 2018 03:42:49 GMT

   svg problem maybe compile gtk-pixbuf2 with --enable-relocations

   https://github.com/Alexpux/MINGW-packages/issues/818

   This is most probably caused by wrong /usr/lib/x86_64-linux-gnu/gdk-pixbuf-2.0/2.10.0/loaders.cache file. Try:
   gdk-pixbuf-query-loaders > $prfx/gdk-pixbuf-2.0/2.10.0/loaders.cache
   to regenerate it.

----
Sun 16 Dec 2018 14:48:21 GMT

   set_thread *is* run just once at the start of ctpl.h.

---
Sun  6 Jan 2019 14:57:30 GMT

   check my blob

   https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/bty626/5055122

---
Sun  6 Jan 2019 15:15:22 GMT

   module opticspy does Zernike moments

---
Tue  8 Jan 2019 14:42:58 GMT

   Here is homebrew's patch for Python

   https://github.com/Homebrew/homebrew-core/blob/master/Formula/python.rb


---
Wed  9 Jan 2019 02:25:27 GMT

   lots of click sounds (just the job?)

   https://freesound.org/people/Jaszunio15/

---
Wed  9 Jan 2019 15:05:52 GMT

   fragment-based folding

   http://www0.cs.ucl.ac.uk/staff/D.Jones/t42morph.html

---
Wed  9 Jan 2019 15:09:38 GMT

   Dan Rigdeon talk:

   quark server: fragment based modelling

   https://zhanglab.ccmb.med.umich.edu/QUARK/

   https://onlinelibrary.wiley.com/doi/full/10.1002/prot.24065

   also

   confold2 (not fragment based)

---
Sun 13 Jan 2019 07:56:23 GMT

  CUDA
  https://www.sharcnet.ca/help/index.php/CUDA_tips_and_tricks#Overview

---
Fri 25 Jan 2019 20:54:54 GMT

Software Violations in Binary Code
https://nixos.org/~eelco/pubs/clones-msr2011-final.pdf