search

hammerlab / cohorts

Utilities for analyzing mutations and neoepitopes in patient cohorts
Apache License 2.0
20 stars 4 forks source link

ExonicSpliceSite with Subsitution/etc. alternate_effect is not included in filtering #255

Closed tavinathanson closed 7 years ago

tavinathanson commented 7 years ago

From @tavinathanson:

:thinking_face: is https://github.com/hammerlab/cohorts/blob/master/cohorts/functions.py#L156 (and others) incomplete because it's not counting ExonicSpliceSite mutations that have Subsitution as the alternate effect?

From @iskandr:

OK, looked at the source and the only properties that forward through the alternate effect are mutant_protein_sequence, modifies_protein_sequence, modifies_coding_sequence So, you have to currently explicitly check, which seems like a recipe for making this mistake often Possible solutions: add a is_substitution method or change how exonic splice sites are handled One possible alternative would be just having every Intragenic effect take a splice_site boolean attribute Or something else that better preserves the meaning of the class hierarchy

From @tavinathanson:

yikes glad we caught it now at least how about we invert the meaning of alternate_effect and have every effect have an optional alternate where ExonicSpliceSite would be one such alternate? similar to splice_site boolean, which also seems fine to me i think either of those is better than is_subsitution which feels like a hack for now i'm going to hack around this in cohorts until we know exactly what we want in varcode