fritzsedlazeck / SURVIVOR

Toolset for SV simulation, comparison and filtering
MIT License
337 stars 46 forks source link

Problems with QV after merge #167

Open BeaSamma opened 2 years ago

BeaSamma commented 2 years ago

Hi @fritzsedlazeck, I'm using SURVIVOR to merge the vcf outputs of 20 individuals. Each vcfs contains the SVs called by 5 different SVs caller (breackdancer, CNVnator, Manta, DELLY and Lumpy). Observing in detailed the QV for each individuals in the first mentioned vcf I noticed that in some cases I don't have a single QV but I have two of them (separed by ","). I report the here the header, the info of the format field and the lines in which I found this.

CHROM POS ID REF ALT QUAL FILTER INFO FORMAT 4051 4187 4212 4567 4573 4586 8391 8440 8657 8807 8811 8816 U1897 U1899 U1902 U1916 U1919 U1920 U1923 U1925

GT:PSV:LN:DR:ST:QV:TY:ID:RAL:AAL:CO

grep -v "#" all_sample_merged.vcf | cut -f10 | cut -f6 -d: | grep -n "," 1001:305,59 1168:194,239 7466:257,413 14255:292,200 16671:198,158 20451:470,584 24476:644,602 28292:215,366

I'd like to produce some QV distribution to understand well how to filter my VCF. Do you have any suggestions on which value I should consider? One of the two values? Both or other?

Thanks for any suggestions, Beatrice