nextgenusfs / funannotate

Eukaryotic Genome Annotation Pipeline
http://funannotate.readthedocs.io
BSD 2-Clause "Simplified" License
301 stars 82 forks source link

Feature Request: an option to funannotate annotate to provide pre-generated BUSCOs #875

Closed docmanny closed 1 year ago

docmanny commented 1 year ago

Currently, funannotate annotate checks to see if an "annotations.busco.txt" file exists before running the custom built-in version of BUSCO. However, newer versions of BUSCO using ODBv10 now have significantly expanded collections of ortholog groups that can be used for annotation. While it is possible to shoe-horn these annotations into funannotate annotate by just manually making this "annotations.busco.txt" file, it would be helpful if the user could provide pre-made BUSCO results to replace the 'run_busco' folder and provide a path to busco/full_table.tsv to the program with a --busco_output option, similar to how you can provide the output file for eggNOG.

Happy to make a PR for this if it would help!

nextgenusfs commented 1 year ago

You can just download the BUSCO v10 models to your FUNANNOTATE_DB folder and use them directly. Keep the _ovb10 in the name but then for --busco_db just give the base name, ie ascomycota will use models in ascomycota_obv10 folder.

docmanny commented 1 year ago

Thank you! I'll try that out. Still, for those of us who are running BUSCO at the end of predict and update to check on the status of the annotations, it might be nice to provide the file to save a bit of computational time, in the case of larger sets like eutheria_odb10 which have a lot of searches to run...

JWDebler commented 1 year ago

I have downloaded the ascomycota_odb10/) dataset and extracted it into the database folder. However, running

funannotate annotate -i funannotate_output --cpus 10 --antismash antismash/Ascochyta_rabiei/Ascochyta_rabiei.gbk --iprscan funannotate_output/annotate_misc/iprscan.xml --busco_db ascomycota

results in

ERROR: ascomycota busco database is not found, install with funannotate setup -b ascomycota

Turns out I had to include the _odb10 in the --busco_db flag for it to work.

funannotate annotate -i funannotate_output --cpus 10 --antismash antismash/Ascochyta_rabiei/Ascochyta_rabiei.gbk --iprscan funannotate_output/annotate_misc/iprscan.xml --busco_db ascomycota_odb10

hyphaltip commented 1 year ago

yes you need to give the full version go the OrthoDB clustered version -- since there was previously _odb9 as a version and will I assume eventually be _odb11