Can't use an undefined value as an ARRAY reference

[tallnuttrbgv]

I have everything working - tested RNAfinder separately and seems ok. When I ran mfannot without RNAfinder (config variable was incorrectly set - I had not renamed DOT to '.'. I got an output but with no tRNAs. Now I ahve RNAfinder working I get the error below:

export EGC=/g/data/nm31/bin/MFannot_data/EGC/
export MFANNOT_EXT_CFG_PATH=/g/data/nm31/bin/MFannot_data/config/
export MFANNOT_MOD_PATH=/g/data/nm31/bin/MFannot_data/models/
export ERPIN_MOD_PATH=/g/data/nm31/bin/MFannot_data/models/Erpin_models/
export MFANNOT_LIB_PATH=/g/data/nm31/bin/MFannot_data/protein_collections/
export PIR_DATAMODEL_PATH=/g/data/nm31/bin/PirObject/PirModels/
export LD_LIBRARY_PATH=/home/554/ta0341/perl5/lib/perl5/
export PERL5LIB=/home/554/ta0341/perl5/lib/perl5/
export PERLLIB=/home/554/ta0341/perl5/lib/perl5/
export RNAFINDER_CFG_PATH=/g/data/nm31/bin/RNAfinder/
export BLASTMAT=/g/data/nm31/db/

mfannot /g/data/dy44/r12.36_organelle_annotation/mt/r12.1_doryanthes_mt.fasta

######################################################################
MFANNOT, ORGANELLAR GENOME ANNOTATION PROGRAM

MFANNOT
MFANNOT: version 1.37
MFANNOT: commit 0990c85
PirModel: commit 81b76f2bd41e82237ea61ed5a03b3a2646eabd8d

Programmed by N. Beck and P. Rioux
######################################################################

---

     Gene Annotation          

---

Parsing masterfile /g/data/dy44/r12.36_organelle_annotation/mt/r12.1_doryanthes_mt.fasta...

1. Translate (flip)...
2. Blast...
3. Parse Blast Results...
4. Select best proteins for Exonerate...
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in ndhK
   Potential frameshift in ndhK
   Potential frameshift in ndhJ
   Potential frameshift in ndhJ
   Potential frameshift in psaA
   Potential frameshift in psaA
5. Annotate genes w/o introns...
6. Intron identification...
7. Annotate genes with introns...
8. Identify gene fusions...
9. Annotate mini exons...
10. Annotate RNA genes...
    'ssrA'...
    Cannot read result for 'ssrA'; maybe there were problems with the commands?
    Cannot read result for 'ssrA'; maybe there were problems with the commands?
    Cannot read result for 'ssrA'; maybe there were problems with the commands?
    'rns'...
    'rnl'...
    'tRNA'...
    'rnpB'...
    'rrn5'...
11. Adjust intron boundaries...

fatal: cannot change to '/mfannot/': No such file or directory
fatal: cannot change to '/mfannot/': No such file or directory
Can't use an undefined value as an ARRAY reference at /g/data/nm31/bin/Mfannot/mfannot line 5746.

[natacha-beck]

I have never seen this issue and tried with a standard genome and found nothing strange. Did you mind to share your genome, I would like to test it ? @tallnuttrbgv

[tallnuttrbgv]

Here it is:
https://drive.google.com/file/d/1zTEO66ElEadOtUJ1RJBof4IkCbptR6QX/view?usp=sharing

[tallnuttrbgv]

I now have a version working in a singularity image that a colleague made for me from the recipe on git:

This had to be run on a VM where I have root (I do not know why)

singularity exec --bind /media/storage/r12.36_organelle_annotation ~/bin/mfannot.sif mfannot mt/r12.1_doryanthes_mt.fasta

Although converting the masterfile output now gives an error I am unsure about. I think it might not be numbering copies of genes properly when they are on separate contigs.

mfannot2gff.pl -m r12.1_doryanthes_mt.fasta.new -g r12.1_doryanthes_mt.gff3
Feature rps10 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature rps10 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad1 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad1 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad1 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad1 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature nad2 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature atp9 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new
Feature atp9 already defined. Please manually verify in r12.1_doryanthes_mt.fasta.new

On my main NCI HPC with a singualrity module, it still does not work:

Error in execution of HMMsearch in RunHMMSearchForLowConservedGenes subroutine
Can't call method "get_resume" on an undefined value at /mfannot/mfannot line 7148.

Is there any chance of changing your terminology for the input file from 'masterfile' to 'input fasta' or similar. It is confusing.

Thanks.

[tallnuttrbgv]

I am trying agat_convert_mfannot2gff.pl to see if that handles multi-copy genes better. What tool do you use to convert to gff3?

Thanks.

[natacha-beck]

For your first issue I suspect an installation, regarding the singularity image you use it is strange that you need to be root, can you please try to pull the singularity image from docker by doing:

singularity pull docker://nbeck/mfannot

The can you test with it.

Regarding mfannot2gff indeed it seems like it did not like the multiple copy of the genes. I personally never convert to gff, please if you found some tools that works well, please let me know.

[tallnuttrbgv]

AGAT has a covnerter that seems to work: agat_convert_mfannot2gff.pl

I tried the new pull of the image and got the error below.

singularity exec --bind /g/data/dy44/r12.36_organelle_annotation ~/bin/mfannot_latest.sif mfannot mt/r12.1_doryanthes_mt.fasta

######################################################################
MFANNOT, ORGANELLAR GENOME ANNOTATION PROGRAM

MFANNOT: version
MFANNOT: commit 104e1c5
PirModel: commit c75c45095cea43c728fa02e83d025b6672aa7b83

Programmed by N. Beck and P. Rioux
######################################################################

---

Gene Annotation

Parsing masterfile mt/r12.1_doryanthes_mt.fasta...

1. Translate (flip)...
2. Blast...
3. Parse Blast Results...
4. Select best proteins for Exonerate...
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in atpA
   Potential frameshift in ndhK
   Potential frameshift in ndhK
   Potential frameshift in ndhJ
   Potential frameshift in ndhJ
   Potential frameshift in psaA
   Potential frameshift in psaA
5. Annotate genes w/o introns...
6. Intron identification...
7. Annotate genes with introns...
8. Identify gene fusions...
9. Annotate mini exons...
10. Annotate RNA genes...
    'rns'...
    'rnl'...
    'tRNA'...
    'rnpB'...
    'rrn5'...
11. Adjust intron boundaries...
12. Identify start codons; identify gene fusions...
13. Find extra genes with HMM
    Error in execution of HMMsearch in RunHMMSearchForLowConservedGenes subroutine
    Error in execution of HMMsearch in RunHMMSearchForLowConservedGenes subroutine
    Error in execution of HMMsearch in RunHMMSearchForLowConservedGenes subroutine
    Error in execution of HMMsearch in RunHMMSearchForLowConservedGenes subroutine
    Error in execution of HMMsearch in RunHMMSearchForLowConservedGenes subroutine

[natacha-beck]

Seems like an issue with some environment variable, can try with --cleanenv option when launching the singularity command.

Unfortunately I am not able to reproduce your issue.

[tallnuttrbgv]

That seemed to work, thanks.