Problematic Fastq qualities with BWA

[chris]

Hi all,

I'm having problems running BWA (0.5.9-r16 and 0.6.2) on some new whole exome data.

The initial alignment runs fine (it seems) but when I want to create a SAM file with 'bwa sampe' it screws up the quality scores. Reading the file with samtools gives this error:

[samopen] SAM header is present: 25 sequences.
Parse error at line 27: sequence and quality are inconsistent

This is the first sequence line of the SAM file:

HISEQ1:382:C14EVACXX:3:1101:1296:1986 83 1 212274224 60 101M = 212274087 -238 AAAGCTGTGGAACGCTACCTCTTCCTTTGAGACCTTGTGGAGAAGGGTCTGAAATGGTAGGCAAAGAGAATAGTTCCCCAGAGAATAAAAACTGGTTGTTG ^_^_$^]^SESC ^\ ^Y# ^_^V^\^W $"#$!#" ^^
^X^^

Compared to an equivalent output from bowtie2:

@PG ID:bowtie2 PN:bowtie2 VN:2.0.0-beta7
HISEQ1:382:C14EVACXX:3:1101:1296:1986 83 1 212274224 42 101M = 212274087 -238 AAAGCTGTGGAACGCTACCTCTTCCTTTGAGACCTTGTGGAGAAGGGTCTGAAATGGTAGGCAAAGAGAATAGTTCCCCAGAGAATAAAAACTGGTTGTTG >>C<2;(8B?>5;6(?CABC@BA?=)7=)7;CIIHEHIGIGC=?BD<@GE>GBB3DEF9AIIIIIGGGGHF?D6E9DHGHCHCBIFF<>FHDDB=:;?1 AS:i:0 XN:i:0 XM:i:0 XO:i:0 XG:i:0 NM:i:0 MD:Z:101 YS:i:0 YT:Z:CP

Anyone else seen this? Is this a problem caused by an incompatibility between BWA and casava 1.8+? I've run BWA in the past with no problems.
TIA

[chris]

Bump. Am I the only one seeing this?

[desaila]

I had similar issues with bwa, switched to bowtie2 like you did

[TiborNagy]

Please write down the command line of BWA. Not just the sampe part, but the aln part also. Usually I get this kind of results when I use -I option in bwa aln.
Other tip: See the next line after the error (in your case line 28). Maybe a newline character can be found on your quality score?

[chris]

I thought this was a dead thread so didn't update.

Problem is/was the '-I' switch in bwa. It doesn't work with newer sanger fastq format files produced with Illumina pipeline v1.8+: remove switch and it works.