multi-threading option doesn't work when we have to use our own reference file

MGHComputationalPathology / CellTics

Center for Integrated Diagnostics at Mass General Hospital NGS tools

BSD 3-Clause "New" or "Revised" License

3 stars 5 forks source link

multi-threading option doesn't work when we have to use our own reference file #7

Open pieterlukasse opened 4 years ago

pieterlukasse commented 4 years ago

The multi-threading option doesn't work when we have to use our own reference file. Any plans to fix this?

alliemclean commented 4 years ago

What error did you get?

alliemclean commented 4 years ago

Is this still a problem?

pieterlukasse commented 4 years ago

hi @alliemclean , yes I believe it is still a problem. I logged it based on a conversation you and my colleague Guruprasad had via email. I'll ask him to add more details here.

guruprasada commented 4 years ago

Hi @alliemclean, @pieterlukasse is referring to our email discussion from September regarding the following two issues: 1) the tool appeared to be querying the UCSC server in spite of us providing our own fasta file via --ref-seq option, 2) the tool appeared to be failing to merge an MNV in the input VCF file with another SNV/indel or MNV. I had sent you my commands and test datasets for these scenarios and I'm happy to resend if needed. Thanks!

alliemclean commented 4 years ago

I just found the email and refreshed my memory. I think we sorted out the first issue. The contig appeared to be missing in the reference so it tried to query UCSC. The MNV's are causing a problem because it does not know how to handle them. I wonder if using vt normalize or vt decompose before using vargrouper would solve this.