BWA sampe fails - invalid BAM binary header (

[Lilach]

Hi,
I am trying to use BWA for analysing human whole exome PE data (Illumina HiSeq) for the first time.
I downloaded the hg19 genome as a reference as indexed it with bwa index.
Then aligned with bwa aln with the following format of command (for eacg fastq file):

nohup bwa aln ~/work_area/hg19_genome/hg19_chromFa.tar.gz SG1177_2_sequence.txt > SG1177_2_aln_sa.sai &

from the ~10Gb fastq files I got ~1.3 Gb sai files.

But when I am trying to use sampe to get the sam files, it doesn't work, and the output file contains the following comments:
[bam_header_read] invalid BAM binary header (this is not a BAM file).
[bam_header_read] invalid BAM binary header (this is not a BAM file).
[bwa_read_seq] the maximum barcode length is 63.
then the list of chromosomes
and end.

Here is the sampe command I used:
nohup bwa sampe ~/work_area/hg19_genome/hg19_chromFa.tar.gz SG1177_1_aln_sa.sai SG1177_2_aln_sa.sai SG1177_1_sequence.txt SG1177_2_sequence.txt &

What did I do wrong?

Many thanks in advance!

[seb567]

Hello,

According to the bwa documentation, the reference should be a FASTA file.


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

Your reference is hg19_chromFa.tar.gz -- this is a compressed archive likely containing many files.

Sébastien Boisvert

Hi,
I am trying to use BWA for analysing human whole exome PE data (Illumina HiSeq) for the first time.
I downloaded the hg19 genome as a reference as indexed it with bwa index.
Then aligned with bwa aln with the following format of command (for eacg fastq file):

nohup bwa aln ~/work_area/hg19_genome/hg19_chromFa.tar.gz SG1177_2_sequence.txt > SG1177_2_aln_sa.sai &

from the ~10Gb fastq files I got ~1.3 Gb sai files.

But when I am trying to use sampe to get the sam files, it doesn't work, and the output file contains the following comments:
[bam_header_read] invalid BAM binary header (this is not a BAM file).
[bam_header_read] invalid BAM binary header (this is not a BAM file).
[bwa_read_seq] the maximum barcode length is 63.
then the list of chromosomes
and end.

Here is the sampe command I used:
nohup bwa sampe ~/work_area/hg19_genome/hg19_chromFa.tar.gz SG1177_1_aln_sa.sai SG1177_2_aln_sa.sai SG1177_1_sequence.txt SG1177_2_sequence.txt &

What did I do wrong?

Many thanks in advance!

[Lilach]

solution

After solving the prolem, I am writing here the solution, in case somebody else will search for a solution for this proble:

The problem was not in giving hg19_chromFa.tar.gz as a reference file. BWA knows to work with it.

The reason for the problem was that I tried to run the bwa commands in the background (in nohup...&) or via a Putty SSH terminal (that crashed in the middle).
This command should NOT be executed in the background, so the only way to run it in a far computer in by VPN. Then, run it without nohup ... &.

[swbarnes2]

Yeah, incompatibility with nohup has been reported on other threads on Seqanswers.

[seb567]

The program screen is useful for that kind of workflow.

You can start your things in a screen, close your window. After that, you can re-ssh to
your box and reconnect to your screen. In one screen, you can have several tabs too.

Sébastien