Added Deposition Date filter

docker-compose.yml

@@ -34,3 +34,5 @@ web:
     - "8000:8000"
   links:
     - mysql
+  environment:
+    - PYTHONUNBUFFERED=0

pgd/settings.py

@@ -4,7 +4,7 @@
 import os
 PROJECT_ROOT = os.path.dirname(os.path.abspath(os.path.dirname(__file__)))
 
-DEBUG = config('DEBUG', default=False, cast=bool)
+DEBUG = True
 TEMPLATE_DEBUG = DEBUG
 
 ADMINS = (

pgd_core/models.py

@@ -19,6 +19,9 @@ class Protein(models.Model):
     # updates allowing up-to-date proteins to be skipped.
     pdb_date    = models.DateTimeField()
 
+    #Deposition date retrieved from the protein file
+    deposition_date = models.DateTimeField(null=True)
+
     # JMT: upgrade to Django 1.6 caused deadlocks on insert
     class Meta:
         select_on_save = True

pgd_search/models.py

@@ -2,6 +2,7 @@
 from math import ceil
 import re
 import cPickle
+import pytz
 
 from django import forms
 from django.conf import settings
@@ -14,6 +15,7 @@
 from pgd_splicer.sidechain import bond_lengths_string_dict, bond_angles_string_dict
 from pgd_core.util import residue_indexes
 
+import datetime
 
 range_re = re.compile("(?<=[^-<>=])-")
 comp_re  = re.compile("^([<>]=?)?")
@@ -148,6 +150,49 @@ def parse_search(self):
         if data.threshold != None:
             query = query.filter(protein__threshold__lte=data.threshold)
 
+        #...filter protein by date
+        if data.depositiondate != None : 
+            range_list = data.depositiondate.split(',')
+            #Setting upper bound in future and lower bound way back in past
+            #This is done if the users doesn't give the initial values
+            lower_bound, higher_bound = 1900, 2020 
+
+            upper_strict, lower_strict = False, False
+            end_year, start_year = None, None
+            for qstring in range_list:
+                qstring = qstring.strip()
+                if '<=' in qstring:
+                    higher_bound = qstring.strip('<=')
+
+                elif '<' in qstring:
+                    higher_bound = qstring.strip('<')
+                    lower_strict = True
+
+                elif '>=' in qstring:
+                    lower_bound = qstring.strip('>=')
+
+                elif '>' in qstring :
+                    lower_bound = qstring.strip('>')
+                    upper_strict = True
+
+            if lower_bound :
+                start_year   = datetime.datetime(int(lower_bound), 1, 1, tzinfo=pytz.timezone('UTC'))
+            if higher_bound :
+                end_year     = datetime.datetime(int(higher_bound), 1, 1, tzinfo=pytz.timezone('UTC'))
+
+            if lower_strict and upper_strict:
+                query = query.filter(protein__deposition_date__lt=end_year,protein__deposition_date__gt=start_year)
+
+            elif lower_strict and not upper_strict :
+                query = query.filter(protein__deposition_date__lt=end_year,protein__deposition_date__gte=start_year)
+
+            elif not lower_strict and upper_strict :
+                query = query.filter(protein__deposition_date__lte=end_year,protein__deposition_date__gt=start_year)
+
+            else :
+                query = query.filter(protein__deposition_date__lte=end_year,protein__deposition_date__gte=start_year)
+
+
         # ...filter by query strings (for values and value ranges)...
         def compare(x,y):
             if x == y:

pgd_search/search/SearchForm.py

@@ -77,6 +77,7 @@ class SearchFormBase(Form):
                                    widget=HiddenInput(attrs={'class':'include'}))
     residues        = ChoiceField(choices=[(i,i) for i in range(1, settings.SEGMENT_SIZE+1)],
                                   initial=3)
+    depositiondate  = CharField(required=False)
 
 # Build a dict for the fields of variable number
 form_dict = {'__module__' : 'pgd_search.views'}

pgd_search/templates/search.html

@@ -93,6 +93,7 @@
                 <td class="label">R-free</td>
                 <td class="field">{{form.rfreeMin}} - {{form.rfreeMax}}</td>
             </tr>
+
             <tr>
                 <td class="label">{{form.threshold.label}}</td>
 

pgd_splicer/ProcessPDBTask.py

@@ -233,6 +233,7 @@ def process_pdb(data):
         protein.rfactor    = float(data['rfactor'])
         protein.rfree      = float(data['rfree'])
         protein.pdb_date   = data['pdb_date']
+        protein.deposition_date = data['deposition_date']
         protein.save()
 
         # 3) Get/Create Chains and save values
@@ -352,7 +353,7 @@ def pdb_file_is_newer(data):
 def amino_present(s):
     """
     Whether a given line contains a valid amino acid.
-    
+
     NB: "SEC" is a valid amino acid, regardless of its absence from
     AA3to1.
 
@@ -407,6 +408,9 @@ def parseWithBioPython(path, props, chains_filter=None):
 
     structure = Bio.PDB.PDBParser().get_structure('pdbname',
                                                   decompressed.name)
+
+    header = Bio.PDB.parse_pdb_header(decompressed)
+    props['deposition_date'] = datetime.strptime(header['deposition_date'], "%Y-%m-%d")
 
     # dssp can't do multiple models. if we ever need to, we'll have to
     # iterate through them
@@ -680,6 +684,9 @@ def parseWithBioPython(path, props, chains_filter=None):
 
         print 'Processed %s residues' % len(residues)
 
+    prop_file = open('props.txt', 'a+')
+    prop_file.write(str(props))
+    prop_file.close()
     return props