bwa question: quality discrepancy between a color-space alignment and its csfastq

[yenhuahuang1]

I am using bwa to perform color-space alignments. I found that the quality line of a mapped read is totally different from its original value in the csfastq file.

==================================
Here is an example:

The record in the sam file:
solid_XX:853_13_240 0 18 48162088 37 48M * 0
0 TCATGAATATAAAACAAGGCCCCTGCAGTCTGGTCCACAGGCCCCCAT VLZ][!!]]]!!TT]ZWSW]]]PO]VOBIRV]UOPQXQEFKNIP]O8
3 XT:A:U CM:i:3 X0:i:1 X1:i:0 XM:i:2 XO:i:0 XG:i:0 MD:Z:6T3T0C36

And the original csfastq record is as follows:
@solid_XX:853_13_240/1
GGCTCGNATTANCACCAGATAAAGCTCGCGGCACGACCCGATAAAACTC
+
=03>>4"7=>:"</<;682<;=>)=851(81<;15269/-023-:<*%%
======================================

I am a bit confused by this discrepancy. Can anyone help? Thanks.

[lh3]

Which version is this? It should be fixed in the latest version. If the latest still produces alignment like this, please let me know. Thanks.

[yenhuahuang1]

bwa 0.5.7 (just downloaded from SF)

[lh3]

I see. I misread the email. The quality in fastq is color quality. The quality in SAM is nucleotide quality. They are different. Please read the bwa paper. This is not a bug, but an essential feature.

[yenhuahuang1]

Many thanks.