#330 Incorrect mapping positions/TLEN field Bowtie2 (2.2.4)

[Felix Krueger]

Hi Ben/Val,

We’ve been encountering situations where the TLEN field appears to do the wrong thing as well, here are two example FastQ entries and the SAM record:
R1.fq
::::::::::::::
@HWI-1KL136:275:C4HL3ACXX:6:2305:6857:43971 1:N:0:TAAGGCGA
CCGTGGCCACCTGTGGGCCACAGCATTCCAGAGGGACTGGCAGCAGAGATGAATACTGCAGTCGTGCCCCTGGTGTGGCCCTGTCTCTTATACACAT
+
DDDA@FFFAEH@BAEAABHCG>?DEEC@B9BFHGE@FE;FB@DFFCCDCH@CHAEHEC:;?;BC5=?A?B;A:<9?@@??B?:@CCCC@>:4@::
::::::::::::::
R2.fq
::::::::::::::
@HWI-1KL136:275:C4HL3ACXX:6:2305:6857:43971 4:N:0:TAAGGCGA
ACACCAGGGGCACGACTGCAGTATTCATCTCTGCTGCCAGTCCCTCTGGAATGCTGTGGCCCACAGGTGGCCACGGGGACCTGTCTCTTATACACAT
+
DDFFHG?FHIGGGHJJIJIJIF@DBHIJJJJJJJJIIJIIHIJIJJFHHDCEEHIEHBEHHFEECECA;@BDDDB@BD@BDDDCCDDDCCDDADCCC

SAM-output
@PG ID:bowtie2 PN:bowtie2 VN:2.2.4 CL:"/bi/apps/bowtie2/2.2.4/bowtie2-align-s --wrapper basic-0 -t -x /bi/scratch/Genomes/Mouse/NCBIM37/Mus_musculus.NCBIM37 -S out.sam -1 R1.fq -2 R2.fq"
HWI-1KL136:275:C4HL3ACXX:6:2305:6857:43971 97 12 114049329 0 80M3I4M4I6M = 114049316 -103 CCGTGGCCACCTGTGGGCCACAGCATTCCAGAGGGACTGGCAGCAGAGATGAATACTGCAGTCGTGCCCCTGGTGTGGCCCTGTCTCTTATACACAT DDDA@FFFAEH@BAEAABHCG>?DEEC@B9BFHGE@FE;FB@DFFCCDCH@CHAEHEC:;?;BC5=?A?B;A:<9?@@??B?:@CCCC@>:4@:: AS:i:-47 XN:i:0 XM:i:4 XO:i:2 XG:i:7 NM:i:11 MD:Z:85G0G0G0C1 YS:i:-50 YT:Z:DP

HWI-1KL136:275:C4HL3ACXX:6:2305:6857:43971 145 12 114049316 0 5M9I3M1D80M = 114049329 103 ATGTGTATAAGAGACAGGTCCCCGTGGCCACCTGTGGGCCACAGCATTCCAGAGGGACTGGCAGCAGAGATGAATACTGCAGTCGTGCCCCTGGTGT CCCDADDCCDDDCCDDDB@DB@BDDDB@;ACECEEFHHEBHEIHEECDHHFJJIJIHIIJIIJJJJJJJJIHBD@FIJIJIJJHGGGIHF?GHFFDD AS:i:-50 XN:i:0 XM:i:2 XO:i:2 XG:i:10 NM:i:12 MD:Z:0G0G6^T80 YS:i:-47 YT:Z:DP

Read 1 is aligns in forward direction and starts at position 114049329. Read 2 is the reverse read and starts at positions 114049316. Yet the TLEN of Read 1 is -103, which leads e.g. SeqMonk to display the read from positions 114049225-114049329 which is most certainly not desired?

Thanks very much for looking into this, Felix