autotrinotate.pl Rnammer

[Alex Martinez]

Hi all,

I am using the autotrinotate.pl script to annotate my transcriptome but I have run into an error that I can't seem to debug. All steps in the automation process (e.g. BLAST, PFAM, tmhmm) are complete aside from Rnammer. Before using the automated perl script, I ran the example rnammer command on the test data to ensure that rnammer was functioning correctly and confirmed this was the case. However, my autotrinotate script keeps failing at the sample place with the following error message: 

CMD: /depot/bioinfo/apps/apps/Trinotate-3.0.1/util/rnammer_support/util/superScaffoldGenerator.pl Trinity_trimmed.fasta transcriptSuperScaffold 100

acc: TRINITY_DN197465_c0_g1_i1

Done.

CMD: perl /group/bioinfo/apps/apps/rnammer-1.2/rnammer -S euk -m tsu,lsu,ssu -gff tmp.superscaff.rnammer.gff < transcriptSuperScaffold.fasta

Error, cmd: perl /group/bioinfo/apps/apps/rnammer-1.2/rnammer -S euk -m tsu,lsu,ssu -gff tmp.superscaff.rnammer.gff < transcriptSuperScaffold.fasta died with ret 256 at /group/bioinfo/apps/apps/Trinotate-3.0.1/util/rnammer_support/RnammerTranscriptome.pl line 80.

While the command successfully produces the .pos.fsa and .pos.fsa.hmmsearchresult files, it seems that rnammer is failing while in the process of creating the .gff file. Has anyone encountered a similar issue during the annotation process? Any thoughts on this issue would be greatly appreciated!

Thanks in advance.

-- 

Alexander Martinez

PhD Student, Ecology and Evolutionary Biology 

Purdue University, Biological Sciences 

Office: Lily G-365

mart...@purdue.edu

[Alex Martinez]

As an update, I have isolated the issue specifically to the command: 

rnammer -S euk -m tsu,lsu,ssu -xml temp.xml -gff temp.superscaff.rnammer.gff < transcriptSuperScaffold.fasta

The script produces the "lsu" xml and .cf files and .hmmsearchresult files. However, it only produces the .xml file for "tsu" but for some reason fails to create the ..cf file. Finally, it produces neither the .xml nor .cf files for "ssu", and there is no .gff file produced at the end. Strangely, when I run the command for the example data it produces the .gff file without issue. Is it possible that file size (i.e. the transcript super scaffold) is an issue? I am curious/hoping that someone else has experienced and worked around a similar issue with rnammer. I would greatly appreciate any insight on this issue.

Thanks in advance,

Alex.

[Brian Haas]

Hi Alex

I hope someone can help.  Rnammer can be a pain to get working, and because of their licensing, I can distribute a better version without writing my own from scratch.

It’s not an essential part of trinotate so don’t let it hold you back.

-Brian

(by iPhone)

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.
To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

[Alex Martinez]

Thanks for getting back to me Brian. I emailed my issue to the developers and am awaiting a response. In the meantime, I've resubmitted the autotrinotate.pl script to run without the RNAMMER component.

Alex.

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

[Alex Martinez]

Just a quick update on Rnammer for those interested. I reached out to CBS asking about the issue outlined above and I was informed that the original Rnammer developer was no longer there and that nobody there had any knowledge about Rnammer functionality. I thought fasta file size might be the issue after getting a few hits using the online Rnammer interface when inputting individual trinity transcripts from my fasta file. I tested this by splitting my fasta file into smaller pieces and running Rnammer on each piece separately, and this ended up giving me several rRNA hits from my Trinity genes. The lesson (I think) is that splitting your .fasta file into more manageable pieces might help if Rnammer is giving you trouble. 

[Brian Haas]

If they would just open up the licensing, I'd be happy to take it over. 

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

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.

To post to this group, send email to trinityrnaseq-users@googlegroups.com.

Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

[Brian Haas]

I'll write them again and see how it goes.  Maybe better luck this time.

To post to this group, send email to trinityrnaseq-users@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas