eastgenomics / eggd_generate_variant_workbook

DNAnexus app for generating xlsx variant workbooks
3 stars 0 forks source link

Reanalysis reads the manifest before getting the given list of panels/genes #41

Closed Yu-jinKim closed 3 years ago

Yu-jinKim commented 3 years ago

Background

TERC was requested in Gemini making it present in the manifest. I thought reanalysis was bypassing the manifest

Issue

https://platform.dnanexus.com/projects/Fx96xBj4zpjP540x7g801JQG/monitor/job/FxZv5Q04zpj93V1b0fjJ3FV2 The log says No transcript for TERC (X012601).

Code causing the issue (lines 126-131)

my %gene_list = readin_manifest( $manifest, $sample);
my %hotspots;

if ( $opts{ 'p' } ) {
  %gene_list = parameter_panels2genes($opts{ 'p' }, $sample);
}

Why

readin_manifest is a subroutine that gets the genes in the manifest and assigns them to transcripts using nirvana_genes2transcripts. But TERC can't be added in g2t at least for now.
$opts{'p'} is the variable containing the panels/genes given for reanalysis.

Why we are only noticing this now

TERC type genes were never requested at the testing level. If it was another gene that we could put in g2t, it could be solved but since this is for TERC...

Proposed fix

my %gene_list;
my %hotspots;

if ( $opts{ 'p' } ) {
  %gene_list = parameter_panels2genes($opts{ 'p' }, $sample);
} else {
  %gene_list = readin_manifest( $manifest, $sample);
}
issue-label-bot[bot] commented 3 years ago

Issue-Label Bot is automatically applying the label bug to this issue, with a confidence of 0.72. Please mark this comment with :thumbsup: or :thumbsdown: to give our bot feedback!

Links: app homepage, dashboard and code for this bot.