There are still redundant variations after "delly filter" step

191 views

Skip to first unread message

邱益彬

unread,

Dec 8, 2020, 9:02:28 AM12/8/20

to delly-users

Dear Tobias,

I have resequencing data for 21 samples, so I applied the "Germline SV calling" pipeline. And apply the "delly filter -f germline" step to remove redundant SV calls. But I found that there was still redundant SV in the final result. These may always happen with imprecise SVs. So should I filter them out?

Here are examples:

14 106235650 DEL00087118 C <DEL> 91 PASS IMPRECISE;SVTYPE=DEL;SVMETHOD=EMBL.DELLYv0.8.5;END=106348294;PE=3;MAPQ=31;CT=3to5;CIPOS=-146,146;CIEND=-146,146;RDRATIO=0.659159;AC=0;AN=2 GT:GL:GQ:FT:RCL:RC:RCR:RDCN:DR:DV:RR:RV 0/0:0,-0.411283,-55.1005:5:LowQual:1889:1025:404:1:10:1:0:0

14 106296869 DUP00087136 C <DUP> 205 PASS IMPRECISE;SVTYPE=DUP;SVMETHOD=EMBL.DELLYv0.8.5;END=106449304;PE=5;MAPQ=41;CT=5to3;CIPOS=-117,117;CIEND=-117,117;RDRATIO=1.41439;AC=0;AN=2 GT:GL:GQ:FT:RCL:RC:RCR:RDCN:DR:DV:RR:RV 0/0:0,-0.903089,-18:10:LowQual:1185:1027:537:1:3:0:0:0

14 106301455 DUP00087137 A <DUP> 129 PASS IMPRECISE;SVTYPE=DUP;SVMETHOD=EMBL.DELLYv0.8.5;END=106589414;PE=3;MAPQ=44;CT=5to3;CIPOS=-175,175;CIEND=-175,175;RDRATIO=1.25674;AC=1;AN=2 GT:GL:GQ:FT:RCL:RC:RCR:RDCN:DR:DV:RR:RV 0/1:-0.296954,0,-9.12521:5:LowQual:3073:2774:3311:1:2:1:0:0

tr

unread,

Dec 10, 2020, 5:31:23 AM12/10/20

to delly-users

Hi,

This depends on how aggressive you merged SVs using delly merge. The 2 key parameters are '-b' and '-r' which adjust the allowed breakpoint deviation and the min. reciprocal overlap for merging. You can also do this at the very end and then simply fetch the genotyped SVs that remain in the merged site list.