BWA - samse

[giverny]

Hi
I am a new user of BWA. I downloaded the 0.5.7.
My aim is to align illumina short reads on the human genome.
First I had problems with bwa samse segmentation fault - as reported by others on this site. Thus I used the program available via MAQ to convert my sequences : fq_all2std.pl
Currently my FASTQ file looks like that that
(...)
@15
NGCANGGCCAGAATGTTTACTCCTTTGGCTCCGTG
+
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%!
@16
NTAGNGCAAAACCATCAATACAAGACTATAGCTGC
+
&,;,&,;;;;98;9;;;;;9;;888;;;9;99;;;!
@17
NCCANCGTCTTGTCTCCGCATACAAGTGGGTCCAT
+
&/6/&/512866647/025266450585)4676%%!
@18
NTTCNCCAGACAGGACAGAAAGGACAGCAGGTGTC
+
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%!
(...)
I hope it's the BWA required format...

My first tests were on a small part of my reads. It worked quickly, without problem.
When I test my complete set, i.e. one file of 5Gb and one other of 10Gb, I'm not sure it works. It is running from Tuesday, and the only line on both .sam files is "[bwa_read_seq] 0.0% bases are trimmed."
Please let me know if it is normal or not. If not what kind of problem have I please. How long does a complete alignment take place (with human reads and genome and without option modification) please?
Thanks a lot for your help

[sperry]

If I'm not mistaken, your quality scores appear to be really low.. I usually convert my Illumina reads to Sanger FASTQ using the 'sol2sanger' feature in the Maq package. You might want to try it out and see how the quality scores compare.

Also, are you using the -t option of 'bwa aln' in order to take advantage of multiple CPUs?

I believe that I was able to align ~7GB of Illumina reads (76bp SE) to the whole human genome in 3-4 hours on an 8-core workstation running Ubuntu Linux.

Hi
I am a new user of BWA. I downloaded the 0.5.7.
My aim is to align illumina short reads on the human genome.
First I had problems with bwa samse segmentation fault - as reported by others on this site. Thus I used the program available via MAQ to convert my sequences : fq_all2std.pl
Currently my FASTQ file looks like that that
(...)
@15
NGCANGGCCAGAATGTTTACTCCTTTGGCTCCGTG
+
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%!
@16
NTAGNGCAAAACCATCAATACAAGACTATAGCTGC
+
&,;,&,;;;;98;9;;;;;9;;888;;;9;99;;;!
@17
NCCANCGTCTTGTCTCCGCATACAAGTGGGTCCAT
+
&/6/&/512866647/025266450585)4676%%!
@18
NTTCNCCAGACAGGACAGAAAGGACAGCAGGTGTC
+
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%!
(...)
I hope it's the BWA required format...

My first tests were on a small part of my reads. It worked quickly, without problem.
When I test my complete set, i.e. one file of 5Gb and one other of 10Gb, I'm not sure it works. It is running from Tuesday, and the only line on both .sam files is "[bwa_read_seq] 0.0% bases are trimmed."
Please let me know if it is normal or not. If not what kind of problem have I please. How long does a complete alignment take place (with human reads and genome and without option modification) please?
Thanks a lot for your help

[giverny]

If I'm not mistaken, your quality scores appear to be really low.. I usually convert my Illumina reads to Sanger FASTQ using the 'sol2sanger' feature in the Maq package. You might want to try it out and see how the quality scores compare.

Also, are you using the -t option of 'bwa aln' in order to take advantage of multiple CPUs?

I believe that I was able to align ~7GB of Illumina reads (76bp SE) to the whole human genome in 3-4 hours on an 8-core workstation running Ubuntu Linux.

Thanks for your answer and sorry for the delay in getting back to you.
Yes the quality of these lines is not the best quality I have on the set... it was just few examples
Finally the problem was relative to the fastq file.
For sure it's more quick now ... and I have results.
Have a good day and thanks again

[GoneSouth]

same problem

Hi guys,

I have exactly the same problem!!
Giverny I would be very greateful if you could describe what was the problem with your fastq file!

best ro

[seq_GA]

Hi, How did you fix the problem of fastq format. I am using maq's sol2sanger program and still get segmentation fault. Please explain. Thanks.

[hoosha]

Hi, did anybody find who to figure out this problem?
i am the same problem but i couldn't find any problem to my fastq files,
thanks

[raela]

I had BWA segmentation fault issues with bwa aln. It turned out my reference fasta file was somehow damaged (I used cat to combine the equine chromosome files into one). Once I received a working genome file, it worked without issues.