Illumina/manta

why my VCF file generated with manta is missing genotype information

Giuseppe1995 opened this issue · 4 comments

Giuseppe1995 commented

Hi, everybody,
I am pretty new to coding and bioinformatics.
I am using Manta as a tool to infer somatic structural variants (SVs) from a paired tumor/normal sample call. However, my somaticSV.vcf.gz file does not contain information about the genotype nor the genotype quality (there is a dot instead of score). I find these fields in the other file (diploidSV.vcf.gz), but I read from the manual that the latter contains infos about germline SVs. My doubt is: which file I should choose for analyzing SVs at somatic level?
Thank you in advance!

hhzhongchn commented

Hello, I also met the same question as you, do you have the answer now?

Giuseppe1995 commented

Hi, yes, now I have an answer to all the doubts raised by myself more than one year ago, now.

  1. For somatic SV detection, is pretty useless to try to infer a genotype, as a tumor is characterized by clonality; what you have to consider is the number of split reads (SR) and paired reads (PR) supporting a somatic SV;
  2. Obviously, the genotype quality cannot be computed if not present a genotype;
  3. For the reason above, if you want to identify SVs which are uniquely present in the tumor you should consider the former file (somaticSV.vcf.gz).
    Hope everything is clear!
YiBenqiu commented

Hi, yes, now I have an answer to all the doubts raised by myself more than one year ago, now.

  1. For somatic SV detection, is pretty useless to try to infer a genotype, as a tumor is characterized by clonality; what you have to consider is the number of split reads (SR) and paired reads (PR) supporting a somatic SV;
  2. Obviously, the genotype quality cannot be computed if not present a genotype;
  3. For the reason above, if you want to identify SVs which are uniquely present in the tumor you should consider the former file (somaticSV.vcf.gz).
    Hope everything is clear!

Thank you for your explanation, but I still don't quite understand how to distinguish which SVs are unique in tomor by RP and SR supporting a somatic SV in the somaticSV.vcf.gz file, I'm sorry, I wonder if you have time to explain it more clearly? Thank you!

YiBenqiu commented

I noticed that in the somaticSV.vcf.gz file, the RP and SR of a somatic SV are different in normal and tumor, but I don't understand what this means.