Problem with --onlyAnn ?

[pmoulos]

Hi,

I am trying to use the pipeline on an already assembled transcriptome (a single FASTA file with ~300k sequences) which I am putting in a directory called onlyAnn as per the instructions. I believe I have followed correctly all the pipeline installation pipelines and tests. However, when I am trying to run:

./nextflow run ~/TransPi/TransPi.nf --onlyAnn -profile conda

I get

N E X T F L O W  ~  version 20.10.0
Launching `/home/moulos/TransPi/TransPi.nf` [happy_koch] - revision: ad8250cb5a
====================================================
  TransPi - Transcriptome Analysis Pipeline v1.0.0-dev
====================================================
TransPi.nf Directory:   /home/moulos/TransPi/TransPi.nf
Launch Directory:       /media/raid/tmp/tmp/genohub/transpi_db
Results Directory:      /media/raid/tmp/tmp/genohub/transpi_db/results
Work Directory:         /media/raid/tmp/tmp/genohub/transpi_db/work
TransPi DBs:            /media/raid/tmp/tmp/genohub/transpi_db
Uniprot DB:             /media/raid/tmp/tmp/genohub/transpi_db/DBs/uniprot_db/uniprot_metazoa_33208.fasta

        Running TransPi with your dataset


        Running only annotation analysis

[-        ] process > transdecoder_longorf     -
[-        ] process > transdecoder_diamond     -
[-        ] process > transdecoder_hmmer       -
[-        ] process > transdecoder_predict     -
[-        ] process > swiss_diamond_trinotate  -
[-        ] process > custom_diamond_trinotate -
[-        ] process > hmmer_trinotate          -
[-        ] process > skip_signalP             -
No such variable: evigene_ch_rnammer

 -- Check script '/home/moulos/TransPi/TransPi.nf' at line: 2299 or see '.nextflow.log' file for more details

Any clues? Please note that I am using a custom path to install TransPi databases instead of the default directory (cloned from GitHub).

Thank you in advance!

[rivera10]

Hello @pmoulos,

Can you pull again the repository? It looks like you may have an old version of TransPi. Currently it is in version 1.1.0-rc and you have v1.0.0-dev. That may solve this issue. Let me know if this still continues.

Best, Ramon

[pmoulos]

Hi @rivera10,

Thanks for the reply! Indeed, the issue regarding the variable name was fixed and the onlyAnn pipeline was able to continue. However, a problem was caused by the fact that the Trinotate authors updated the help scripts to reflect new download links for Gene Ontology files, so I think that the pipeline crashed because it was not able to build the initial Trinotate SQLite boiler plate. I was able to finish manually based on the reported commands. Maybe you can consider updating Trinotate somehow or maybe this has to do with the respective conda channel?

Thanks again!

[rivera10]

Hello @pmoulos,

Thanks for letting me know about the update of Trinotate. I will check and update the links.

Best, Ramon