A samtools v0.1.11 bug???

[nturay]

Hi all,

I've extracted one paired-end read from BWA aligned result and use samtools v0.1.11 to convert them to a BAM file as my test data set as below :

HWUSI-EAS1812:1:23:18626:4628#0 99 chr1 10542 10 76M = 10817 314 CGAAATCTGGGCAGAGGAGAACGCAGCTCCGCCCTCGCGGGGCTCTCCGGGTCTGTGCTGAGGAGAACGCAACTCC ############################################################################ XT:A:U NM:i:3 SM:i:10 AM:i:10 X0:i:1 X1:i:3 XM:i:3 XO:i:0 XG:i:0 MD:Z:9T8C21T35 XA:Z:chr15,-102520467,76M,4;chr12,-95006,76M,4;chr16,+60483,76M,4;

HWUSI-EAS1812:1:23:18626:4628#0 147 chr1 10817 10 6S24M2D7M1D5M34S = 10542 -314 TCCAGGGGGGTGGAGGCGTGGCGCAGGCGCAGAGGCGGCCGCCTCCGGGTGGGGGTTGGTGGGGCGTGTGGTGCAG ###################################################@14@9-700014>+1B7BB:4??-? XT:A:M NM:i:3 SM:i:10 AM:i:10 XM:i:0 XO:i:2 XG:i:3 MD:Z:24^AG7^C5

Then I use Bio-SamTools v 1.27 to grab the chr, start, end, strand information from this BAM file.

The result gives as below :

chr1 10542 10617 1
chr1 10817 10855 -1

chr1 10817 10855 -1
chr1 10542 10568 1

The first one (Blue) seems to be the correct answer, but its mate pair gives different ending position.

I am wondering whether it's a samtools 0.1.11 (r851) bug or so?

Ray