[xinwu]

I am using bwa Version: 0.5.8 (r1442) on CentOS, a machine with lots of RAM.

I downloaded the unmasked mouse genome from

http://hgdownload.cse.ucsc.edu/goldenPath/mm9/bigZips/chromFa.tar.gz

I then created an index for bwa like this:

bwa index -a bwtsw mm9.fa

I then aligned my illumina short reads agains the genome and get sai files. My goal is bed files, so I want to convert sai -> sam -> bam -> bed to get there.

When I try to convert from sai to sam like this:

bwa samse SOMEPATH/mm9.fa stuff.sai stuff.fq > stuff.sam

I get:
convert to sequence coordinate... Segmentation fault

But, I have done the very same procedure with different illumina short read sequences (even higher amount of data) for a different organism (rat), and it went OK.

Any help would be very much appreciated. (I am afraid I am not able to patch bwa, though)

I run into the same error with you when I use another genome. No ideas...
Another BWA run at background over 70 hours when generate SAM file from SAI file for transcriptome.

[hrajasim]

segmentation fault - solved for me

Folks, I came to this thread looking for an answer to the segmentation fault that I too encountered a couple weeks ago. However, I did not find an answer then - Upon investigation, here is what I found:

The bwa aln command when run with a "nohup" and &, inserted the bwa output trace such as:

[bwa_aln_core] convert to sequence coordinate... 0.43 sec
[bwa_aln_core] refine gapped alignments... 0.29 sec
[bwa_aln_core] print alignments...

into the sai file itself (rather than to nohup.out). WHen I deleted all lines containing these bwa_aln_core tags and the 0.29 sec from the files, it worked just fine. I came back to this thread to share this finding - hope it helps.

[DrDolittle]

I can confirm the report by harjasim (thanks for the hint!) trying to align Illumina 40 bp single end to the rat genome. I got two different errors, however: 1) "bwa samse" stalled at the exact same position, 2) most recently "bwa samse" complained about missing "*fa.nt.ann" files. I cannot recall why this changed, I didn't install new versions and had both errors on two different systems. The only settings I changed were the -n and -t options. Maybe Heng Li can advise on this or patch bwa?

Cheers
Top of *.sai file
[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... 603.08 sec
[bwa_aln_core] write to the disk... ^C^@^@^@^K^@^@^@^D^@^@^@^C^B^@^@^E^@^@^@

used versions
cutadapt version 0.9
bwa version 0.5.9-r16
samtools version 0.1.12a (r862)
Mac OS X 10.6.6/Red Hat Enterprise Linux Server release 5.4 (Tikanga)

[fali]

Folks, I came to this thread looking for an answer to the segmentation fault that I too encountered a couple weeks ago. However, I did not find an answer then - Upon investigation, here is what I found:

The bwa aln command when run with a "nohup" and &, inserted the bwa output trace such as:

[bwa_aln_core] convert to sequence coordinate... 0.43 sec
[bwa_aln_core] refine gapped alignments... 0.29 sec
[bwa_aln_core] print alignments...

into the sai file itself (rather than to nohup.out). WHen I deleted all lines containing these bwa_aln_core tags and the 0.29 sec from the files, it worked just fine. I came back to this thread to share this finding - hope it helps.

I am also having this problem. However, I am a beginner. How can I enter the sai file and delete these lines? Thanks!

[DrDolittle]

I assume, you're running it with the nohup / background option? If you will start it without this, you shouldn't run into any problem. I hadn't at least. You could also try to just delete those lines with a texteditor, but the files are large! I have never tried this, but harjasim seemed to have gone that route.

[nilshomer]

I am also having this problem. However, I am a beginner. How can I enter the sai file and delete these lines? Thanks!

Try piping the stderr stream too. So "nohup bwa aln ... > out.sai 2> out.stderr &".

[variadevang]

is a 24GB RAM machine alright for aligning the .sai files using bwa sampe?????????

[carolW]

segmentation fault with bwa samse

Hi,
When I try to execute the following command:

../pgm/bwa-0.7.3a/bwa samse -f SRR062641.filt.sam -r "@RG\tID:Exome1\tLB:Exome1\tSM:Exome1\t:PL:ILLUMINA" ../hg19/Homo_sapiens/UCSC/hg19/Sequence/BWAIndex/version0.6.0/genome.fa SRR062641.filt.sai SRR062641.filt.fastq

I just get Segmentation fault (core dumped)

The data is 1000genomes data 255 bp, 109811 sequences.

Where did I do wrong?

Thanks,

Carol

[mastal]

bwa samse segmentation fault

I'm not sure what is causing your segmentation fault, but if your reads are 250 bp long, you should be using the newer bwa mem algorithm. bwa backtrack (bwa aln, bwa samse) is designed for reads up to 100 bp.

[carolW]

The number of reads comes from the output of bw aln (see below)

How to specify the index file with bwa mem?

Thanks
--------------------------------------------------------
[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... 114.72 sec
[bwa_aln_core] write to the disk... 0.04 sec
[bwa_aln_core] 109811 sequences have been processed.
[main] Version: 0.6.1-r104
[main] CMD: bwa aln -t 4 -f SRR062641.filt.sai -I ../../hg19/Homo_sapiens/UCSC/hg19/Sequence/BWAIndex/version0.6.0/genome.fa ../../data/SRR062641.filt.fastq
[main] Real time: 111.442 sec; CPU: 119.167 sec

[mastal]

bwa samse segmentation fault

How to specify the index file with bwa mem?

I think it's the same as for bwa aln.
Have a look at the bwa manual page:

http://bio-bwa.sourceforge.net/bwa.shtml

[carolW]

well, it doesn't work. I have used different combinations. The problem is that there is no parameter for the index file and on the command line, bwa mem can't recognize it.

../pgm/bwa-0.7.3a/bwa mem samse SRR062641.filt.sam -r "@RG\tID:Exome1\tLB:Exome1\tSM:Exome1\t:PL:ILLUMINA" ../hg19/Homo_sapiens/UCSC/hg19/Sequence/BWAIndex/version0.6.0/genome.fa SRR062641.filt.fastq SRR062641.filt.mem.sai
[E::bwa_idx_load] fail to locate the index files

[GenoMax]

well, it doesn't work. I have used different combinations. The problem is that there is no parameter for the index file and on the command line, bwa mem can't recognize it.

../pgm/bwa-0.7.3a/bwa mem samse SRR062641.filt.sam -r "@RG\tID:Exome1\tLB:Exome1\tSM:Exome1\t:PL:ILLUMINA" ../hg19/Homo_sapiens/UCSC/hg19/Sequence/BWAIndex/version0.6.0/genome.fa SRR062641.filt.fastq SRR062641.filt.mem.sai
[E::bwa_idx_load] fail to locate the index files

Correct order for the files for samse (as indicated in the manual page):

bwa samse ref.fa aln_sa.sai short_read.fq > aln-se.sam

Also this line from the same manual:

Alignment algorithms are invoked with different sub-commands: aln/samse/sampe for BWA-backtrack, bwasw for BWA-SW and mem for the BWA-MEM algorithm.

[carolW]

I have been using the parameters that were given in the tutorial

http://seqanswers.com/wiki/How-to/ex...ina_technology

If I use in the order that you indicated, I will get segmentation fault again

../pgm/bwa-0.7.3a/bwa samse ../hg19/Homo_sapiens/UCSC/hg19/Sequence/BWAIndex/version0.6.0/genome.fa SRR062641.filt.sai SRR062641.filt.fastq > SRR062641.filt.sam
Segmentation fault (core dumped)

[GenoMax]

How much memory do you have on the machine you are using? Does the segmentation fault happen right away or after some time?