[maker-devel] substr outside of string in PhatHits_utils.pm

Tue Apr 17 09:58:16 MDT 2018

It runs fine to completion for me (on both 3.01.01 and 3.01.02). Since I’m using your output, no external tools are called, it just parses the reports already written in the directory and finishes.

This suggests that any issue is either with your version of Perl or a component MAKER is using (such as BioPerl). I am using Perl 5.16.3 and BioPerl 1.007002 (the CPAN version).

Note if you are using BioPerl live or the GitHub release it still shows version 1.007002 but will not necessarily match the CPAN version as the GitHub version counter does not get iterated with each commit. So make sure you are not accidentally using BioPerl live from GitHub (only use CPAN or let MAKER do the install of BioPerl if it’s not system wide). Also you are using Exonerate 2.4 instead of the stable 2.2 release. That shouldn’t make any difference since I am just parsing your output that is already in the folder and not running exonerate. But it may be worth an off chance look.

Finally what you may want to do is download a new version of Perl, then install that and run MAKER with that version just to make sure Perl or something installed inside your Perl is not generating thee issue.

I would try the current stable release of perl (since it comes all ready to go - no compiling needed). Alternatively you can also try perlbrew to get a specific version (but it will have to compile against local libraries).

—Carson

> On Apr 17, 2018, at 6:54 AM, Jacques Dainat <jacques.dainat at nbis.se> wrote:
> 
> I tried with the last version 3.01.02-beta, still the same error.
> 
> I agree that it could be something wrong with the input GFF3 file, but I don’t find what it could be.
> 
> I have loaded the whole folder as user guest_5602.
> 
> I’m looking forward to hearing from you.
> 
> /Jacques
> 
>> On 16 Apr 2018, at 22:11, Carson Holt <carsonhh at gmail.com <mailto:carsonhh at gmail.com>> wrote:
>> 
>> I can’t replicate your failure. Normally this error indicates there is something wrong with the input GFF3 file.
>> 
>> If you want, you can run this on your own machine to generate the failure, then tarball up the complete maker directory for the failure and upload it here —> http://weatherby.genetics.utah.edu/cgi-bin/mwas/bug.cgi <http://weatherby.genetics.utah.edu/cgi-bin/mwas/bug.cgi>
>> 
>> Also try the most current version of MAKER (3.01.02 - October 2017). See if it happens for you there.
>> 
>> —Carson
>> 
>> 
>> 
>> 
>>> On Apr 13, 2018, at 6:52 AM, Jacques Dainat <jacques.dainat at nbis.se <mailto:jacques.dainat at nbis.se>> wrote:
>>> 
>>> Dear Carson,
>>> 
>>> I come back to you still with the same problem: substr outside of string at /sw/bioinfo/maker/3.01.1-beta-OMPI/bin/../lib/PhatHit_utils.pm line 850.
>>> Since our last conversation in January I have seen in the MAKER mailing list that one more person (seoanezonjic) had this issue.
>>> In January, the only way I found to avoid the problem was to remove the gff files that were related to the issue.
>>> 
>>> I have again the problem for a new annotation project. On the 6 `EST` gff files I’m using (produced in the same way, with Stringtie and converted in gff3 alignment style), 2 of them are raising the error.
>>> To try to better get where the problem come from I have minimise the tools used within MAKER. So no repeat masking and abinitio tools activated, only protein in fasta format and one EST file in gff format.
>>> Using only the gff est file with est2genome=1 works
>>> Using only protein in fasta with protein2genome works
>>> Using the gff EST file and the protein in fasta format with  protein2genome and est2genome or only protein2genome doesn’t work 
>>> 
>>> The problem occurs when protein alignments try to be extended by the EST information.
>>> I tried using the same tool versions as you (BioPerl 1.007002, BLAST+ 2.7.1, Exonerate 2.2.0) but still the same problem.
>>> One of the interesting thing is that the problem does not occur when I used the protein in gff format.
>>> 
>>> Here is one gff file raising the error. If you want to try in the same conditions you will have to use/dowmload the swissprot database (uniprot rewieved only).
>>> 
>>> 
>>> <9529.1.136329.CGATGT.gff3>
>>> <scaffold_1017.fa>
>>> 
>>> I hope we will finish to find a solution to this problem…
>>> 
>>> Best regards,
>>> 
>>> Jacques
>>> -------------------------------------------------
>>> Jacques Dainat, Ph.D.
>>> NBIS (National Bioinformatics Infrastructure Sweden)
>>> Genome Annotation Service
>>> http://nbis.se/about/staff/jacques-dainat/ <http://nbis.se/about/staff/jacques-dainat/>
>>> http://nbis.se <http://nbis.se/>
>>> 
>>> Address:
>>> Uppsala University, Biomedicinska Centrum
>>> Department of Medical Biochemistry Microbiology, Genomics
>>> Husargatan 3, box 582
>>> S-75123 Uppsala Sweden
>>> Phone: +46 18 471 46 25
>>> 
>> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://box290.bluehost.com/pipermail/maker-devel_yandell-lab.org/attachments/20180417/76deecda/attachment.html>