HaplotypeCaller shifts deletion CIGARs in -bamout

HI,
I have noticed on the number of occasions that the cigar strings of deletions in the haplotype file generated with -bamout are incorrectly shifted by 1 or 2 basepairs. It does not affect variant calling (calls in vcf fie are correct). Looks like a bug? See below the IGV screenshot (top panel - haplotype bam, middle - realigned recalibrated bam, bottom - vcf file) and below are a couple of reads from bam files:

From haplotype bam:
HISEQ:410:C1UVYACXX:2:1107:15486:59236 99 1 11740287 54 41M12D59M = 11740353 166 AGGTGATCCGCCCGCCTCGGCCTCCCAAAGTGCTAGGATTACAGGCGCCCGACCAAAGAATTTTTAAAAGTCCACCCTGTACCCACTCTGTGGTTCTGGG @CACFEGHECGEEBECEG>;DFDGFDCEDDFDGEFDFCCEDDEFBDFAFDEBDGG5555555555555555555555555555555555555555555555 HC:i:476358318 MC:Z:100M BD:Z:MMPOPSQRPOOPLNNONMONOONMNKOLCMOKOOMNONNMLNKPOOMNOKMOONOLCMMMMMDDDLLCCMOONOKOLOPLOOOLPLPOOQNNRRQRNPPL RG:Z:130809_SN172_0410_BC1UVYACXX_TGACCA_L002_R1_001 BI:Z:PPQRRTSSQRRQMQQPRPQPRPRPPMSOGQRORSQRPQQOPQNRPRPQPNRSSRTPHRPQRPJJJQOHHRSSQTPROTUQTSSOUQTSTVRSUUTWRTRM NM:i:12 MQ:i:60 AS:i:82 XS:i:47

Same read from realigned recalibrated bam:
HISEQ:410:C1UVYACXX:2:1107:15486:59236 99 1 11740287 54 43M12D57M = 11740353 166 AGGTGATCCGCCCGCCTCGGCCTCCCAAAGTGCTAGGATTACAGGCGCCCGACCAAAGAATTTTTAAAAGTCCACCCTGTACCCACTCTGTGGTTCTGGG @CACFEGHECGEEBECEG>;[email protected]GHFHJGGDGFFDFFBEE MC:Z:100M BD:Z:MMPOPSQRPOOPLNNONMONOONMNKOLCMOKOOMNONNMLNKPOOMNOKMOONOLCMMMMMDDDLLCCMOONOKOLOPLOOOLPLPOOQNNRRQRNPPL RG:Z:130809_SN172_0410_BC1UVYACXX_TGACCA_L002_R1_001 BI:Z:PPQRRTSSQRRQMQQPRPQPRPRPPMSOGQRORSQRPQQOPQNRPRPQPNRSSRTPHRPQRPJJJQOHHRSSQTPROTUQTSSOUQTSTVRSUUTWRTRM NM:i:12 MQ:i:60 AS:i:82 XS:i:47

Thank you,
Dalia

Tagged:

Best Answer

Answers

Sign In or Register to comment.