convert GATK germline CNV .vcf output to .seg/.cns format to plot

Question

[cocchi.e89]

I have some CNV calls operated through GATK germline CNV pipeline, the output of such pipeline is a .vcf file (.segments.vcf.gz) that looks like this:

zgrep -v "^#" /home/ec3408/germlineCNV/A13661_PostGermlineCNVCaller/A13661.segments.vcf.gz | head 
chr1    821001  CNV_chr1_821001_1632000 N   .   3076.53 .   END=1632000 GT:CN:NP:QA:QS:QSE:QSS  0:2:405:58:3077:105:92
chr1    1665001 CNV_chr1_1665001_1684000    N   <DUP>   51.10   .   END=1684000 GT:CN:NP:QA:QS:QSE:QSS  1:4:3:51:51:57:51
chr1    1753001 CNV_chr1_1753001_12903000   N   .   3076.53 .   END=12903000    GT:CN:NP:QA:QS:QSE:QSS  0:2:6171:25:3077:4:95
chr1    12903001    CNV_chr1_12903001_12904000  N   <DEL>   3.53    .   END=12904000    GT:CN:NP:QA:QS:QSE:QSS  1:1:1:4:4:4:4
chr1    12904001    CNV_chr1_12904001_12906000  N   <DEL>   45.26   .   END=12906000    GT:CN:NP:QA:QS:QSE:QSS  1:0:2:6:45:6:45
chr1    12906001    CNV_chr1_12906001_12912000  N   .   57.81   .   END=12912000    GT:CN:NP:QA:QS:QSE:QSS  0:2:5:15:58:5:6
chr1    12912001    CNV_chr1_12912001_12913000  N   <DEL>   5.09    .   END=12913000    GT:CN:NP:QA:QS:QSE:QSS  1:1:1:5:5:5:5
chr1    12913001    CNV_chr1_12913001_12918000  N   <DEL>   41.30   .   END=12918000    GT:CN:NP:QA:QS:QSE:QSS  1:0:3:18:41:18:40
chr1    12919001    CNV_chr1_12919001_12922000  N   <DEL>   32.01   .   END=12922000    GT:CN:NP:QA:QS:QSE:QSS  1:1:2:25:32:43:18
chr1    12922001    CNV_chr1_12922001_12924000  N   <DEL>   135.11  .   END=12924000    GT:CN:NP:QA:QS:QSE:QSS  1:0:2:72:135:99:72

is there any way to convert this to a .seg or a .cns file, in order to plot etc. with CNVkit, IGV etc.

Thank you very much in advance for any help!

Answer 1

[newbio17]

There is a way to do this in their documentation here under section "Visualize gCNV results in IGV and compare to truth set calls". It's a three step process right below the IGV figures.

Once you have the .seg files, you can use cnvkit.py import-seg to convert them into .cns files.