samtools mpileup unexpected behaviour

[CHRYSES]

Hi,

The command below:
samtools mpileup -ugf 454AllContigs.fasta my.bam | bcftools view -bvcg - > var.raw.bcf


exits with the following error:

[mpileup] 2 samples in 1 input files
<mpileup> Set max per-sample depth to 4000
[bcf_sync] incorrect number of fields (7 != 5) at 43:101



Would that mean that the bam file is missing fields? It is created via the standard methods as outlined in the BWA and samtools manuals.

The alignments are made using bwa bwasw where 454 reads served as query and references.

The RG fields are created according to the samtools manual.

The following fields have been set while merging:
@RG ID:dUMMY1 SM:dUMMY1 LB:454 PL:454
@RG ID:dUMMY2 SM:dUMMY2 LB:454 PL:454
(...)


Does anyone have an idea what the error cause is? If so, please share it here.


The description 7 != 5 is not helping much to track the reason.

Thanks,

Best regards

[lh3]

This has not happened for a while. Could you try the latest version from SVN? What if you use "mpileup -I" to disable indel calling. If both fail, do you have a small example for me to try? Thanks.

[CHRYSES]

Originally Posted by lh3

This has not happened for a while. Could you try the latest version from SVN? What if you use "mpileup -I" to disable indel calling. If both fail, do you have a small example for me to try? Thanks.

The proposed solution with "mpileup -I" has indeed avoided the crash+error, but at the same time it renders samtools to a snp caller only. No indels are called.

I had done an svn checkout of the trunk lately. The version I am using is:

$ svn info
Path: .
URL: https://samtools.svn.sourceforge.net...samtools/trunk
Repository Root: https://samtools.svn.sourceforge.net/svnroot/samtools
Repository UUID: fd675618-d7d1-4fc3-a54f-e58cce934862
Revision: 902
Node Kind: directory
Schedule: normal
Last Changed Author: lh3lh3
Last Changed Rev: 902
Last Changed Date: 2011-01-24 03:46:18 +0100 (Mon, 24 Jan 2011)


The samtools binary shows this:
$ samtools

Program: samtools (Tools for alignments in the SAM format)
Version: 0.1.12-10 (r896)


Note that I could not use GATK for indel calling in 454 reads so I wanted to give samtools a shot in this, but it seems like it cannot handle it (yet).

Do you have any suggestions on making the samtools work with this kind of data?

Thanks.

Regards

[Seq84]

The proposed solution with "mpileup -I" has indeed avoided the crash+error, but at the same time it renders samtools to a snp caller only. No indels are called.

Me too i had the same problem of CHRYSES and i've solved with -I option.

Why it happens? is a version problem (0.1.12-10 (r896-r862)) or is a problem internal to the .bam file?

Thanks in advance

[cnoirot]

I still have a problem with the version 0.1.18 (r982:295)
The reference base is always set to N but in fasta file it's not the case !
I can't perform varFilter.
Do you have ever seen this problem ?