BWA crashes in Fedora

[sdvie]

Let me reply to my own question:

What happened is that the solid2fastq script obviously did not process the reads completely (as a part of the batch process).
Funnily enough, the last read of the fastq file being incomplete did not prevent bwa aln to run and generate a .sai file.
Only in the bwa samse step, this caused trouble and led to said error message.

Hope this helps.

cheers,
Sophia

[sdvie]

Dear all,

I got this error message, too, when running bwa samse (bwa-0.5.9rc1):

[bwa_aln_core] refine gapped alignments... Segmentation fault (core dumped)

however, the difference to the previous posts is that
1. my original SOLiD reads were reformatted using bwa's solid2fastq.pl alright, and
2. I ran a batch of 12 samples that were processed all the same way. Only 7 gave this error, 5 were processed without error.

Any ideas why this happens will be appreciated.

thanks,
Sophia

[ulz_peter]

Thank you for your help.
I was just too stupid for reading the manual. I did new indexing of the genome with the -a is argument which is not recommended for large genomes. After applying the -a bwt option in indexing, alignment worked.

[idonaldson]

I had the same problem (BWA on Fedora 13) and gave up on it for a long time. Then someone passed me a BLOG link:

http://kevin-gattaca.blogspot.com/2010/08/howto-do-bwa-mapping-in-colorspace.html

My problem seemed to be i was using a solid2fastq script from a different program (BFAST i think. Why was i doing this, because it is C-compiled therefore much faster!). Once i used solid2fast from the BWA package it started working.

I hope this helps!

[ulz_peter]

Lately I changed OS from Ubuntu to Fedora (KDE) (both 64bit).
While in Ubuntu BWA worked perfectly it now crashes while trying to align a 7.4GB Illumina fastQ file which I could align perfectly in Ubuntu.

The output is

[bwa_aln] 17bp reads: max_diff = 2
[bwa_aln] 38bp reads: max_diff = 3
[bwa_aln] 64bp reads: max_diff = 4
[bwa_aln] 93bp reads: max_diff = 5
[bwa_aln] 124bp reads: max_diff = 6
[bwa_aln] 157bp reads: max_diff = 7
[bwa_aln] 190bp reads: max_diff = 8
[bwa_aln] 225bp reads: max_diff = 9
[bwa_aln_core] calculate SA coordinate... Segmentation fault (core dumped)

I use 16GB quadcore 64bit Fedora 14 and I tried every bwa version from 0.5.6 to 0.5.9.

The command was:

bwa aln ~/RefSeq/hg19_070510/Chromosomes/hg19.fa test.fq > test.aln

Any help appreciated (looking in your direction lh3...)