search

PacificBiosciences / HiFiCNV

Copy number variant caller and depth visualization utility for PacBio HiFi reads
BSD 3-Clause Clear License
35 stars 4 forks source link

question about using deepvariant or deepsomatic to get vcf file as HiFiCNV input #28

Closed DayTimeMouse closed 2 months ago

DayTimeMouse commented 3 months ago

Hi,

I see the vcf is generated from deepvariant.

Is it more reasonable to use deepsomatic to call snv in tumors, as opposed to using deepvariant?

I am confused about the choice of vcf, should I use germline or somatic SNV?

Best wishes.

holtjma commented 3 months ago

Hi @DayTimeMouse,

I see the vcf is generated from deepvariant. Is it more reasonable to use deepsomatic to call snv in tumors, as opposed to using deepvariant?

Let me start by just point out that HiFiCNV is designed for germline samples only, it does not support tumor which may have mosaic copy numbers.

I am confused about the choice of vcf, should I use germline or somatic SNV?

As of now, the provided VCF file is only used to generate an auxiliary data track, {OUTPUT_PREFIX}.{sample_name}.maf.bw. It will not influence the CNV calling components (i.e., you should get duplicate CNV calls regardless of the provided VCF file).

Given the above, I think the choice is largely up to you. If it was me and I was trying to glean information from a tumor sample, I would probably still use the deepvariant file because it will be calling mostly germline variants and the MAF track will be most informative for those "normal" variants. Low frequency somatic variants will not be particular informative in the CNV MAF track.

Matt

holtjma commented 2 months ago

Closing due to inactivity.