Entering edit mode
2.6 years ago
Sbrillo
▴
10
Hi,
I'm using bowtie1 to align unpaired reads (fastq) to Pacbio reads (fasta) in an HPC. I divided the reference (Pacbio reads) into many chunks in order to reduce the timing of the alignment.
This are the command that i'm using but is not working:
bowtie-build $TMPDIR/${PBS_ARRAY_INDEX}_m_pac.fasta $TMPDIR/${PBS_ARRAY_INDEX}_m_pac
bowtie -a -t -5 0 -3 5 -p 32 -v 0 -x $TMPDIR/${PBS_ARRAY_INDEX}_m_pac -q $TMPDIR/m.fastq --suppress 1,2,4,5,6,7,8,9 $TMPDIR/male.txt 2> /rds/general/user/mv419/ephemeral/redkmer-droso/data-Y/pacBio_illmapping/logs/${PBS_ARRAY_INDEX}_male_log.txt
The bowtie command is not finding the indexing file in TMPDIR.
Any suggestion?
i might have made a mistake in the bowtie command.
Thanks in advance
Are you sure
$TMPDIR
is not being cleared the moment yourbowtie-build
is done? Try using a non TMP dir maybe?Thanks for the feedback.
I thought about this but when I run the following command on bowtie (is an old version command that is not completely supported in the latest bowtie releases):
bowtie is correctly reading the index in TMPDIR, however, this command is not working properly on all the chunks and sometimes is not generating a complete output file.
What do you think about this command ? i might be close to a solution.
I'm not familiar with the bowtie command, just with HPC environments. If it's a problem with bowtie versions, maybe ensure that the bowtie-build is the same version (or a compatible version to) the bowtie executable.