MPileup SNP and indel caller (Galalxy Version 0.0.1) error

Question: MPileup SNP and indel caller (Galalxy Version 0.0.1) error

9 months ago by

vanam • 30

United States

vanam • 30 wrote:

Hi, I'm trying to rerun an MPileup job I successfully ran before, using MPileup SNP and indel caller (Galaxy Version 0.0.1). I'm using the 'Run this job again' icon (the icon with two arrows that make a circle), and am not changing any of the settings or the input file. The job is now failing with the following error:

[E::bcf_hdr_read] Invalid BCF2 magic string: only BCFv2.2 is supported python: ��: Unknown error 74404736

Is there something wrong with the tool? I'm using Main.

Thanks, Marcy

software error mpileup samtools • 386 views

ADD COMMENT • link •

modified 9 months ago • written 9 months ago by vanam • 30

9 months ago by

Nate Coraor ♦ 3.2k

United States

Nate Coraor ♦ 3.2k wrote:

Although old versions of tools should not stop working, I did notice that the inputs were generated with a very old version of Picard, and this is a very old version of MPileup. Do you get the same results with the newest versions of these tools?

ADD COMMENT • link written 9 months ago by Nate Coraor ♦ 3.2k

9 months ago by

vanam • 30

United States

vanam • 30 wrote:

Hi Nate, I haven't tried the newest versions. I need to rerun an old pipeline. Will it be possible to fix the older MPileup version or do I need to move to the newer versions? Thanks, Marcy

ADD COMMENT • link written 9 months ago by vanam • 30

I'm looking in to whether this can be fixed. In general we make a strong commitment to keep things that once worked, working forever after. But this is not always possible to do, especially for tools that predate the more robust dependency management that we've added over the past few years.

ADD REPLY • link written 9 months ago by Nate Coraor ♦ 3.2k

Thank you very much for looking into this, Nate! -Marcy

ADD REPLY • link written 9 months ago by vanam • 30

Hi Nate, I'm trying to move to the newer versions, but when I use the same data with MPileup 2.0 on Main a line is generated in the resulting vcf for every base, with <x> as the ALT, resulting in over a million lines (MPileup 0.0.1 used to call about 450 variants on this targeted sequencing data). For the few lines where an actual variant is called (about 450-500 lines), the ALT is instead A,<x> (or C,<x> etc.). Do you have any idea how I might resolve this? I see a couple of other posts with the same issue (https://biostar.usegalaxy.org/p/16026/ and https://biostar.usegalaxy.org/p/18282/ but it doesn't look like these other issues ever got resolved. Thanks very much for any help you can provide! Marcy

ADD REPLY • link written 6 months ago by vanam • 30

The tool options and behavior can change between versions. Use the most current version: Mpileup 2.1.3. Review Set advanced options to make sure the filtering/reporting is what you want.

This output is from Samtools itself (that is what the second Q&A is about, also here https://www.biostars.org/p/161981/#162000) and everyone has had to come up with a workaround to deal with the new data output, not just those working with the tool at Galaxy Main https://usegalaxy.org.

I made a request to add bcftools call to https://usegalaxy.org https://github.com/galaxyproject/usegalaxy-playbook/issues/123

Similar posts • Search »