Errors with Platypus

199 views
Skip to first unread message

Ehi

unread,
Sep 19, 2018, 2:59:48 AM9/19/18
to Platypus Users
Greetings

Thanks for your work on this software. I have been using Platypus for the better part of a year as part of the Fast-GBS pipeline, for calling variants from a genotyping by sequencing samples. I have processed ~300 samples through this pipeline without issues (in one batch). 

Recently I have been having problems running Platypus. I recently got a batch of ~200 samples from the same sequencer as used for the prior samples, but this time I cannot get Platypus to run on them. It always fails for all regions and skips all regions. The error says “Error was Something was screwy here”; see snippet below.

 

I used Picard tools to test the BAM files I am using and they do not seem different from the samples that worked. I found that the BAMs are missing Read Groups (@RG) but that was not different from the BAM files that worked in the past. I tried using subsets of my data and found that running the last ~60 samples avoided the issue, but using other subsets of 60 did not help. I am pretty lost on what to try next. I would like to get data from all the samples. Do you have any suggestions or know what is causing that weird error message?

 

Thanks for the help!




sdf.jpg



Andy Rimmer

unread,
Sep 19, 2018, 4:43:05 AM9/19/18
to Ehi, Platypus Users
Hi,

I think this is actually a bug. Platypus uses the RG tags to determine the sample names,  and how many samples are present in each BAM. If there are no RG tags present, Platypus uses the name of the BAM file as the sample name. This could cause problems, if you have several BAM files with the same names (in different directories, presumably). Is this possible? If so, the easy solution is to re-name the BAMs.

Kind regards,
Andy


--
You received this message because you are subscribed to the Google Groups "Platypus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to platypus-users+unsubscribe@googlegroups.com.
To post to this group, send email to platypus-users@googlegroups.com.
Visit this group at https://groups.google.com/group/platypus-users.
To view this discussion on the web, visit https://groups.google.com/d/msgid/platypus-users/5b28cdfd-6919-4d42-864b-b7909c7a31f8%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--
Dr Andrew (Andy) Rimmer

Ehi

unread,
Sep 19, 2018, 1:02:11 PM9/19/18
to Platypus Users

Thanks for the suggestion Andy!

I am sure that all the samples are named differently because the data is only stored in a single directory. I think some of the samples were resequenced and merged into one fastq file. Do you think that this can trick Platypus into thinking there are more samples than there really are, causing errors? Do this mean I have to correct the @RG header? 

 

Ehi Akhirome


On Wednesday, September 19, 2018 at 3:43:05 AM UTC-5, Andy Rimmer wrote:
Hi,

I think this is actually a bug. Platypus uses the RG tags to determine the sample names,  and how many samples are present in each BAM. If there are no RG tags present, Platypus uses the name of the BAM file as the sample name. This could cause problems, if you have several BAM files with the same names (in different directories, presumably). Is this possible? If so, the easy solution is to re-name the BAMs.

Kind regards,
Andy
On Wed, Sep 19, 2018 at 7:59 AM, Ehi <akhi...@wusm.wustl.edu> wrote:
Greetings

Thanks for your work on this software. I have been using Platypus for the better part of a year as part of the Fast-GBS pipeline, for calling variants from a genotyping by sequencing samples. I have processed ~300 samples through this pipeline without issues (in one batch). 

Recently I have been having problems running Platypus. I recently got a batch of ~200 samples from the same sequencer as used for the prior samples, but this time I cannot get Platypus to run on them. It always fails for all regions and skips all regions. The error says “Error was Something was screwy here”; see snippet below.

 

I used Picard tools to test the BAM files I am using and they do not seem different from the samples that worked. I found that the BAMs are missing Read Groups (@RG) but that was not different from the BAM files that worked in the past. I tried using subsets of my data and found that running the last ~60 samples avoided the issue, but using other subsets of 60 did not help. I am pretty lost on what to try next. I would like to get data from all the samples. Do you have any suggestions or know what is causing that weird error message?

 

Thanks for the help!




sdf.jpg



--
You received this message because you are subscribed to the Google Groups "Platypus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to platypus-user...@googlegroups.com.
To post to this group, send email to platypu...@googlegroups.com.

Ehi

unread,
Sep 20, 2018, 5:47:21 PM9/20/18
to Platypus Users
Hi Andy,

Any thoughts on what the issue could be?

Ehi Akhirome

Andy Rimmer

unread,
Sep 21, 2018, 4:21:37 AM9/21/18
to Ehi, Platypus Users
I think you need to fix the RG header tags. If they are present, but incorrect, it will cause problems. If there is no RG header tag, Platypus will assume one sample per BAM, and will use the BAM file name as the sample name (you will see this in the VCF output)
Andy


To unsubscribe from this group and stop receiving emails from it, send an email to platypus-users+unsubscribe@googlegroups.com.
To post to this group, send email to platypus-users@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Ehi

unread,
Dec 5, 2018, 5:33:38 PM12/5/18
to Platypus Users
I don't think it is the problem because all of my files have unique names since they are all in the same directory. If the RG tags are missing ans defaulting to the file name, I deduce that this would not cause these errors.
Any other thoughts? I can send you a few of these files and maybe you can quickly run it and see? Let me know if you can. It would be a tremendous help to me since I have been struggling with this for months.

Thanks,
Ehi Akhirome
Reply all
Reply to author
Forward
0 new messages