No output from samtool's mpileup

[mikheyev]

I am trying to run samtools for SNP detection, but neither mpileup or the deprecated pileup produces any output.

My data are in SAM format, which I convert to BAM using samtools thusly:

./samtools view -bSt FAIFILE.fai INFILE.sam > OUTFILE.bam

No errors, seem to get a functional BAM file.

Now, following the manual,

./samtools mpileup -ugf REFSEQ.fasta INFILE.bam | bcftools view -bvcg - > var.raw.bcf

This produces the following output to sdterr, and no output to screen
[mpileup] 1 samples in 1 input files
<mpileup> Set max per-sample depth to 8000

I have aslo tried a similar approach with pileup:

./samtools pileup -vcf REFSEQ.fasta INFILE.bam

This computes for LONG time, but still produces no output.

I have no idea what is going wrong... Please help!

Sasha

[kopi-o]

I had the very same problem, and found this:

http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTiktWy6kjS8ng3qphopVApLi1H_Fd1NkOm8yaHSf%40mail.gmail.com&forum_name=samtools-help

Not sure it will help you, but it did help me because I had exactly such a bam file (paired-end reads where the read pairs had not been matched up, due to some issues with upstream software.)

[swbarnes2]

I also tried to run the command like that, and I got the same thing. It didn't write anywhere.

So I don't pipe. I run mpileup with -ugf to make a file, (and that step can take a very long time) and then I feed that file to bcftools to make the bcf once it's done.

The second step is so much faster than the first that you don't save much time by piping.

[mikheyev]

Thanks a lot for both suggestions. However, when I tried downgrading, all I got was a memory error, and no joy.
Not piping with -ugf gives the following output, but no output file.

[mpileup] 1 samples in 1 input files
?BCa??BCF#chromosom1chromosome2chromosome3 s7_2.bam%##samtoolsVersion=0.1.13 (r926:134)
<mpileup> Set max per-sample depth to 8000

I checked to see whether there was some problem with my SAM to BAM conversion by uploading my test file to Galaxy and running mpileup there, and everything seemed to work. So, could there be some weirdness in the way samtools are compiled on my system? I run Mac OS, but received no errors during compilation.

[kopi-o]

Too bad.

You could try to trawl the Samtools mailing list archives (e g http://sourceforge.net/mailarchive/f...=samtools-help) for further clues.

[shabhonam]

I guess, you need to sort your bam file. i had similar problem and I got away with it by simply sorting it.

[mmartin]

Relevant for me was lh3's comment in another thread:: Use the -A option with mpileup in order to include non-properly-paired reads.