[maker-devel] ERROR: Failed while processing the chunk divide!!

Fri Feb 14 12:09:08 MST 2014

Hi Hossein, 

So, this is what is going on. The problem is with the GFF3 file, and the problem is that the exon features in that GFF3 should have the mRNA as their parent instead of the gene. When you deleted the "-mRNA-1", the Name of the mRNA became the same as the Name of the gene, which restored the proper relationship between the features. The same problem exists for the CDS features.

The solution for this is to make the exon and CDS parent's "point" to the mRNA and not the gene. Since MAKER has very regular rules for making names, this should be pretty straight forward. You should be ok with just adding "-mRNA-1" to the end of all the exon and CDS lines. This will work unless there some mRNAs with alternative splice forms because then the mRNA's will end with something like "-mRNA-2". 

I've attached a script that should do this for you. 

Run it with this command

"perl fix_gff3_script.pl <your_gff3> > <fixed_gff3>"

And then run MAKER with the fixed gff3 file in place of the old gff3 file. 

Let me know if that works, 

Daniel

Daniel Ence
Graduate Student
Eccles Institute of Human Genetics
University of Utah
15 North 2030 East, Room 2100
Salt Lake City, UT 84112-5330
________________________________________
From: Borhan, Hossein [Hossein.Borhan at AGR.GC.CA]
Sent: Thursday, February 13, 2014 3:27 PM
To: Daniel Ence
Subject: Re: ERROR: Failed while processing the chunk divide!!

Dear Daniel

I downloaded maker 2.31 and ran the same scaffold. Again it gave error on
the gff file. I then removed the word mRNA-1 from my gff file and ran it
again. It seems to have worked this time. Attached are std error files for
first try std-err (the one that failed) and 2nd one named std-err-wo-mRNA
(that apparently worked).  Since the gff file is as evidence only I
thought it should not matter to remove the mRNA-1 naming form the gff file.

Cheers

HB

On 14-02-12 12:59 PM, "Daniel Ence" <dence at genetics.utah.edu> wrote:

>Hi Hossein,
>
>So, after looking at the gff3 and your control files, I had an idea.
>There's the part of the control file called "Re-annotation Using MAKER
>Derived GFF3", but you can also passthrough features from a gff3 using
>the "est_gff", "protein_gff", "rm_gff", "pred_gff", "model_gff" lines.
>
>Sometimes we encounter problems with the MAKER passthrough. Could you try
>dividing the gff3 file into the different feature sources and passing it
>through the "est_gff" etc options and not with the MAKER passthrough?
>That will tell us if the problem is with the gff3 file or with how MAKER
>is processing it.
>
>Another also to check is to make sure that the contig names in the gff3
>file match the contig names in the fasta file that you're annotating.
>
>Thanks,
>Daniel
>
>
>
>Graduate Student
>Eccles Institute of Human Genetics
>University of Utah
>15 North 2030 East, Room 2100
>Salt Lake City, UT 84112-5330
>________________________________________
>From: Borhan, Hossein [Hossein.Borhan at AGR.GC.CA]
>Sent: Wednesday, February 12, 2014 8:49 AM
>To: Daniel Ence
>Subject: Re: ERROR: Failed while processing the chunk divide!!
>
>Dear Daniel
>
>
>I have generated the files that you requested. I choose Sc00009 from my
>genome which is 30 kb and was one of the scaffolds coming up with error.
>In addition to Ctl files and error output file I also attached a part of
>the gff file related to SC00009 that is indicated in the error message.
>
>
>Thanks for helping with this
>
>
>
>Regards
>
>
>HB
>
>
>
>
>
>
>
>
>
>
>
>
>On 14-02-11 4:59 PM, "Daniel Ence" <dence at genetics.utah.edu> wrote:
>
>>Hi Hossen,
>>
>>I think that what would be the most help right now is if you ran MAKER on
>>only one of those contigs that are failing and send me the entire error
>>output along with the maker control files that you are using. It looks
>>like the error is coming from the gff3 files that you are using as input.
>>
>>Thanks,
>>Daniel
>>
>>
>>
>>Daniel Ence
>>Graduate Student
>>Eccles Institute of Human Genetics
>>University of Utah
>>15 North 2030 East, Room 2100
>>Salt Lake City, UT 84112-5330
>>________________________________________
>>From: Borhan, Hossein [Hossein.Borhan at AGR.GC.CA]
>>Sent: Tuesday, February 11, 2014 3:51 PM
>>To: Daniel Ence
>>Subject: ERROR: Failed while processing the chunk divide!!
>>
>>Dear Daniel
>>
>>I re-started maker and it is still running. But in error our file that
>>has
>>been generated so far it seems that smaller conitgs are affected. There
>>are contigs of 2-4 kb with this error but also I noticed a contig of 30kb
>>length having this error
>>
>>I was wondering if I need to change the setting in the maker_opt file
>>
>>#-----MAKER Behavior Options
>>max_dna_len=100000 #length for dividing up contigs into chunks
>>(increases/decreases  memory usage)
>>min_contig=1 #skip genome contigs below this length (under 10kb are often
>>useless)
>>
>>
>>If I understand correctly max_dna_len   divide conitgs  of over 100kb to
>>smaller chucks. However it is not clear to me that for the min_contig
>>option if the default contig length is 10kb or less, then why I have
>>error
>>message for 30kb long contigs. Should I change this to 0
>>
>>Here is an example of the error message for one of the contigs
>>
>>
>>#--------- command -------------#
>>Widget::exonerate::est2genome:
>>/usr/local/exonerate-2.2.0-x86_64/bin/exonerate  -q
>>/raid01/projects/Plasmodiophora/brassicae/PT3/version2/Maker-config/P.bra
>>s
>>s
>>icae.PT3.v1.genome.maker.output/P.brassicae.PT3.v1.genome_datastore/35
>>/17/PbPT3Sc00001//theVoid.PbPT3Sc00001/comp14545_c0_seq1.fasta
>>-t
>>/raid01/projects/Plasmodiophora/brassicae/PT3/version2/Maker-config/P.bra
>>s
>>s
>>icae.PT3.v1.genome.maker.output/P.brassicae.PT3.v1.genom
>>e_datastore/35/17/PbPT3Sc00001//theVoid.PbPT3Sc00001/PbPT3Sc00001.235-113
>>6
>>.
>>fasta
>>-Q dna -T dna --model est2genome
>>--minintron 20 --showcigar --percent 20 >
>>/raid01/projects/Plasmodiophora/brassica
>>e/PT3/version2/Maker-config/P.brassicae.PT3.v1.genome.maker.output/P.bras
>>s
>>i
>>cae.PT3.v1.genome_datastore/35/17/PbPT3Sc00001//theVoid.PbPT3Sc00001/PbPT
>>3
>>S
>>c00001.235-1136.comp14545_c0_seq1.est_exonerate
>>#-------------------------------#
>>cleaning blastn...
>>cleaning tblastx...
>>cleaning blastx...
>>ERROR: Failed on
>>PbPT3Sc00001_S_0.8_1-mRNA-1
>>Check your input GFF3 file for errors!
>>(from GFFDB)
>>
>>FATAL ERROR
>>ERROR: Failed while processing the chunk
>>divide!!
>>
>>ERROR: Chunk failed at level 17
>>!!
>>FAILED CONTIG:PbPT3Sc00001
>>
>>
>>
>>
>>--Next Contig--
>>
>>
>>
>>
>>
>>
>>Regards
>>
>>
>>HB
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>On 14-02-11 12:37 PM, "Daniel Ence" <dence at genetics.utah.edu> wrote:
>>
>>>Hossein,
>>>
>>>Ok. So since this error came up on a local install, I'm going to need
>>>some more information to understand what went wrong. Is it the same
>>>contig that always causes this error? If it is, then is the the only
>>>error or warning that MAKER encounters while running on this contig? Or,
>>>if multiple contigs fail, then is it always the same error?
>>>
>>>If you can narrow it down to the smallest possible dataset that
>>>consistently gives the same error, then we canb egin to understand
>>>what's
>>>wrong.
>>>
>>>Thanks,
>>>Daniel
>>>
>>>
>>>Daniel Ence
>>>Graduate Student
>>>Eccles Institute of Human Genetics
>>>University of Utah
>>>15 North 2030 East, Room 2100
>>>Salt Lake City, UT 84112-5330
>>>________________________________________
>>>From: Borhan, Hossein [Hossein.Borhan at AGR.GC.CA]
>>>Sent: Tuesday, February 11, 2014 11:20 AM
>>>To: Daniel Ence
>>>Subject: Re: [maker-devel] Falied to create new account
>>>
>>>Hi Daniel
>>>
>>>I running it through the local server at my work
>>>
>>>
>>>
>>>
>>>
>>>
>>>M. Hossein Borhan, Ph.D.
>>>Research Scientist/ Chercheur Scientifique
>>>Saskatoon Research Centre/Centre de Recherches de Saskatoon
>>>Agriculture and Agri-Food Canada/Agriculture et Agroalimentaire Canada
>>>107 Science Place, Saskatoon, SK.,S7N 0X2
>>>Telephone/Téléphone: (306) 385-9441
>>>Facsimile/Télécopieur: (306) 385-9482
>>>Hossein.borhan at agr.gc.ca
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>On 14-02-11 12:16 PM, "Daniel Ence" <dence at genetics.utah.edu> wrote:
>>>
>>>>Hi Hossein,
>>>>
>>>>Did you encounter this error while you were running MAKER on your local
>>>>machine or through the MAKER web annotation service?
>>>>
>>>>Thanks,
>>>>Daniel
>>>>
>>>>
>>>>Daniel Ence
>>>>Graduate Student
>>>>Eccles Institute of Human Genetics
>>>>University of Utah
>>>>15 North 2030 East, Room 2100
>>>>Salt Lake City, UT 84112-5330
>>>>________________________________________
>>>>From: Carson Holt [carsonhh at gmail.com]
>>>>Sent: Tuesday, February 11, 2014 10:18 AM
>>>>To: Daniel Ence
>>>>Cc: Mark Yandell
>>>>Subject: FW: [maker-devel] Falied to create new account
>>>>
>>>>Hey Daniel could you download his dataset, and see if you can replicate
>>>>the error.  Also check if this was an MWAS job or a local maker run
>>>>(his
>>>>dataset will already be there for MWAS, you just need the job ID).
>>>>
>>>>Thanks,
>>>>Carson
>>>>
>>>>On 2/11/14, 10:16 AM, "Borhan, Hossein" <Hossein.Borhan at AGR.GC.CA>
>>>>wrote:
>>>>
>>>>>Hi Carson
>>>>>
>>>>>
>>>>>I encountered this error while running maker
>>>>>
>>>>>FATAL ERROR
>>>>>ERROR: Failed while processing the chunk divide!!
>>>>>
>>>>>ERROR: Chunk failed at level 17
>>>>>!!
>>>>>FAILED CONTIG:PbPT3Sc00006
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>HB
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>
>>
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: fix_gff3_script.pl
Type: application/octet-stream
Size: 349 bytes
Desc: fix_gff3_script.pl
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20140214/364c961e/attachment-0003.obj>