Bowtie output files

[igarciacuevas]

Hi!

I am trying to replicate some biocomputational procedures from a scientific paper related to ChIP-Seq data. When using bowtie as the alligner and introducing the output (specified as .sam format) in the MACS software it tells me that the file format is not recognized.
However, after converting the unrecognized sam file to bam format using samtools, MACS does recognize the file format and starts analyzing the data.

Could anyone shed some light on this?

My bowtie and macs commands are:

bowtie -m 2 -q -p 24 --sam hg19 File1.fastq File1.sam

macs14 callpeak -t File1.sam -c ControlFile.sam
-n ExperimentName -S -g hs --pvalue 1e-9 --keep-dup=auto



Iñigo

[westerman]

Using the --format command in MACS will insist on the input files being in a given format. It is possible that by using the default automatic format detection that MACS got confused by your file.

That is my best guess and solution.

[blancha]

You should switch to MACS2 anyway.
MACS2 might recognize the format correctly.
Otherwise, I would specify the format explicitly as westerman suggested, and also check the header and the first entries of the SAM file for any potential problems with the file.

You're also using 24 threads during the alignment with Bowtie. If you don't have that many processors available, the output can be mixed up. Of course, if you do have the 24 processors available on your machine, then you should be using the 24 threads to do the alignment quickly.