search

hartwigmedical / hmftools

Various algorithms for analysing genomics data
GNU General Public License v3.0
193 stars 59 forks source link

PURPLE: structural_vcf and sv_recovery_vcf #47

Closed vladsavelyev closed 5 years ago

vladsavelyev commented 5 years ago

I'm wondering what's the difference between structural_vcf and sv_recovery_vcf arguments? In previous versions I used to feed my manta output using structural_vcf, which if I understand correctly was used both for guidance of CN breakpoint calling and for SV recovery. If I read the code correctly, these 2 purposes are now separated. Should I still feed the same VCF with both arguments, or there are some specific recommendations?

p-priestley commented 5 years ago

Please check readme.

https://github.com/hartwigmedical/hmftools/tree/master/purity-ploidy-estimator#7-structural-variant-recovery

Note It is mainly designed with GRIDSS SV calling in mind. GRIDSS produces a high confidence (similar to Manta + BPI) and low confidence call set. It is optional, so I would probably not recommend using it if you don't use GRIDSS

vladsavelyev commented 5 years ago

Thanks for the explanation! I read that bit but wasn't sure about the difference between those 2 arguments. All clear now.

We are still using Manta+BPI in production, but switching to GRIDSS now. Thanks for your work on PURPLE and all recent improvements.