SURVIVOR: SURVIVOR merging to wrong svtypes.

Hi Again,

I running lattes version (1.0.7) of SURVIVOR and found strange behavior of merging, though i am using svtype into consideration.

list: BreakDancerSortVCF.vcf dellySort.vcf lumpySort.vcf MantaCandidateSV…vcf.gz

##OPTIONS Caller = 3 Type No minimum sv length No strand info

SURVIVOR merge list 1000 3 1 0 0 30 MergeOut.vcf

chr15	78630477	MantaINS~364025~1~2~0~0~0	TG	TTTGGTGTACCACCCAGAGACCACAGAGACAGTTGGATGCCATTCTTCACTCACTGAAGTAGAGAAATGTATGTATA	.	PASS	SUPP=3;SUPP_VEC=0111;SVLEN=585;SVTYPE=DUP;SVMETHOD=SURVIVOR1.0.7;CHR2=chr15;END=78631125;CIPOS=0,621;CIEND=-27,0;STRANDS=+-	GT:PSV:LN:DR:ST:QV:TY:ID:RAL:AAL:CO	./.:NaN:0:0,0:--:NaN:NaN:NaN:NAN:NAN:NAN	0/1:NA:554:18,0:-+:.:DUP:DUP00014037:NA:NA:chr15_78630545-chr15_78631099	./.:NA:554:0,32:+-:.,.:DUP,DEL:27945:NA:NA:chr15_78630544-chr15_78631098,chr15_78630605-chr15_78631099	./.:NA:648:0,0:+-:.,.,.:DUP,INS,INS:MantaINS~364025~1~2~0~0~0:TG:TTTGGTGTACCACCCAGAGACCACAGAGACAGTTGGATGCCATTCTTCACTCACTGAAGTAGAGAAATGTATGTATA:chr15_78630477-chr15_78631125,chr15_78631098-chr15_78631099,chr15_78631098-chr15_78631099

You may see SURVIVOR took call of DUP and DEL from “delly” as well as “DUP”, “INS”, “INS”, “INS” from Manta.

I really dont get it, since INS in both case have either one base difference or none (start and end is same in case of manta). It should NOT merge INS.

Example of DEL in all four vcf (random not the above example):

MantaCandidateSV.vcf
chr1	3009350	MantaDEL:7:0:0:0:0:0	TTGTGTGTGTGTG	T	.	.	END=3009362;SVTYPE=DEL;SVLEN=-12;CIGAR=1M12D;CIPOS=0,41;HOMLEN=41;HOMSEQ=TGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGT;UPSTREAM_PAIR_COUNT=0;DOWNSTREAM_PAIR_COUNT=0;PAIR_COUNT=0

GT190533_FKDL190729564-1a_H225YCCX2_cat_L2_1.fq.gz_dedupe_BreakDancerSortVCF.vcf
chr1	3499556	.	.	.	.	PASS	PROGRAM=breakdancer;SVTYPE=DEL;SVLEN=2855;SVEND=3502300	GT:DP	1/.:16

GT190533_FKDL190729564-1a_H225YCCX2_cat_L2_1.fq.gz_dedupe_dellySort.vcf
chr1	3031359	DEL00000000	T	<DEL>	0	PASS	PRECISE;SVTYPE=DEL;SVMETHOD=EMBL.DELLYv0.8.1;CHR2=chr1;END=3031717;PE=0;MAPQ=0;CT=3to5;CIPOS=-8,8;CIEND=-8,8;SRMAPQ=60;INSLEN=0;HOMLEN=7;SR=16;SRQ=0.994845;CONSENSUS=CATGCTCTATACCTAATACCCACTTATCAGTGAGTATACACCATGTGTGATATTTTGTGATGGAGTTACCTCACTCAGGATATTATTTTGTAGATCCATTCATTTGCCTGTGAGTTTCATGAAGTCACTATTTTTAATAGCTGAGTAATACTCCATTTTGTAAATATATCAGATTTTCTGTATCTATTCCTCT;CE=1.89153	GT:GL:GQ:FT:RCL:RC:RCR:CN:DR:DV:RR:RV	1/1:-103.199,-7.82553,0:78:PASS:5534:22:5773:0:0:0:0:26

GT190533_FKDL190729564-1a_H225YCCX2_cat_L2_1.fq.gz_dedupe_lumpySort.vcf
chr1	3031359	1	N	<DEL>	.	.	SVTYPE=DEL;STRANDS=+-:19;SVLEN=-359;END=3031718;CIPOS=0,0;CIEND=0,0;CIPOS95=0,0;CIEND95=0,0;SU=19;PE=0;SR=19	GT:SU:PE:SR	./.:19:0:19

You may notice that there is no SVLEN for Delly whereas lumpy and manta have negative SVLEN while breakdancer have +ve SVLEN. is that make confusion to SURVIVOR ?

Regards, Rupesh