Ignore gap sequences and reporting of repeats in canonical form

[olekto]

When running ULTRA I sometimes get Ns as a tandem repeat, or included as a larger pattern. Is this intentional?

I would prefer that all gap bases (Ns) are ignored in the output. Is this possible?

Also, it is possible to report the repeats in canonical form? For instance, GA, AG,TC, CT are all represented by the repeat type AG. This is from Chambers, G. K., and E. S. MacAvoy. 2000. (https://www.sciencedirect.com/science/article/pii/S0305049100002339#APPA) which states: "That microsatellite satellite repeat sequence be represented by the simplest and most alphabetical formula possible, e.g. –(CA)15– rather than –(TG)15– or alternatives."

Is this possible?

Thank you.

Sincerely,
Ole

[DanielOlson]

Hi Ole,

To clarify, is the ULTRA-labeled repetitive sequence pure Ns (this would not be intentional), or is the the repetitive pattern pure Ns, but the repetitive sequence is a somewhat poorly conserved repetitive pattern? Right now, low consensus characters in the repetitive label are reported as N, but I could see that could be somewhat confusing when all characters in the repetitive label has low-consensus . Perhaps a better behavior would be to call repeat labels with a large amount of Ns low-complexity sequence. Filtering low-complexity sequence or repetitive labels with lots of Ns is not currently supported, but it seems that it should be. I'll mark this as a feature to be developed.

Right the repeat labeling convention from Chambers, G. K., and E. S. MacAvoy. 2000 is not supported, although it seems like it should be. I'll mark this feature for development also.

It will likely be a few weeks until I get the next version of ULTRA out, although if this has some urgency for you then I'll try to get to it sooner.

Thanks and best wishes,
Daniel

[olekto]

Thank you, Daniel.

I was not completely aware that Ns could be part of the consensus. Looking at the output now, I see that this seems to be working fine in most cases.

However, I am talking about cases like this:
SUPER_12 16792543 16807954 100 -544.245850 NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN 1 0 NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN

Where it looks like the whole region is just Ns, and where I know there's a gap. I see that the score is -544.245850, so I guess I could have /should have filtered it due to that.

Thank you for offering to getting a version of ULTRA out sooner. However, I process the output with some Python scripts, so I get it to where I want it anyhow.

Ole

[DanielOlson]

Yikes - that certainly should not be reported as a repeat. Is there any chance you could email me the FASTA file and the runtime parameters you are using? You can contact me at danielolson@arizona.edu.

Thanks for bringing this to my attention, and sorry for any inconvenience!

Daniel