Opened 4 weeks ago

Closed 4 weeks ago

Last modified 3 weeks ago

#19110 closed enhancement (fixed)

Make mutationscores associate easier to use when sequences don't match

Reported by: Tom Goddard Owned by: Tom Goddard
Priority: moderate Milestone:
Component: Structure Analysis Version:
Keywords: Cc: Elaine Meng
Blocked By: Blocking:
Notify when closed: Platform: all
Project: ChimeraX

Description

Willow said associating structures with DMS data where the sequences don't quite match (e.g. mouse vs human opioid receptor) is difficult in ChimeraX. There is a command to do it

mutationsscores structure #1 alignSequences ABCG2_HUMAN

It requires the sequence of the DMS data is specified because their may be incomplete data. And if you have lots of structures with several chains if you specify all of them it associates with chains that are not at all similar.

To improve this would be nice to allow aligning to the sequence derived from the DMS data maybe with option "alignSequences true". Also add another option minimumSequenceIdentity defaulting to say 50% that rejects an associations that don't have a good sequence alignment.

Change History (2)

comment:1 by Tom Goddard, 4 weeks ago

Cc: Elaine Meng added
Resolution: fixed
Status: assignedclosed

Done.

Made mutationscores structure command able to align to mutation data sequence with X replacing any missing residues instead of a specified sequence using "alignSequences true". Formerly the alignSequences option required specifying a sequence or an alignment. Now a boolean value is also allowed.

Also added minimumPercentIdentity option (default 50) to mutationscores structure command. This means that 50% of the mutation data residues have to be matched with identical amino acid type for the mutationscores structure command to associate a structure with the mutation data.

Also made alignment use Needleman-Wunsch computed on the user's computer instead of Clustal Omega web service.

The allowMismatches option defaults to true if the alignSequences option is specified, otherwise it defaults to false.

comment:2 by Elaine Meng, 3 weeks ago

Great, sounds very useful!  Here is an attempt at revised documentation

<https://www.cgl.ucsf.edu/home/meng/chimerax/vdocs/user/commands/mutationscores.html#structure>

I guess one question is whether the "chains" option of the "open" command still behaves as currently documented or whether it too has become fancier:

<https://www.cgl.ucsf.edu/home/meng/chimerax/vdocs/user/commands/open.html#chains>
Note: See TracTickets for help on using tickets.