Download Databases

[tpellegrinetti]

Description of the bug

Hi,

Thank you very much for developing this amazing tool.

I was trying to use funcscan on my FASTA dataset, but I encountered issues with automatically downloading the databases.

The error I received is: [nf-core/funcscan] error: HMM model files not found for --amp_hmmsearch_models! Please check input.

The command that I runned:
nextflow run nf-core/funcscan -r 1.1.5 -profile conda --input samplesheet.csv --outdir results --run_amp_screening --run_arg_screening --run_bgc_screening --run_taxa_classification --save_db

I am unsure if it's necessary to download all the databases manually or if the pipeline is supposed to handle this automatically. I would really appreciate any help with running it successfully.

Thank you!
Thierry

Command used and terminal output

Relevant files

No response

System information

No response

[jasmezz]

Hi @tpellegrinetti,
sorry for late reply!

Easy answer to your question: Either run your command with --amp_skip_hmmsearch or run the latest release which fixed that (i.e. -r 2.0.0).

Longer explanation: All databases are downloaded automatically, yes. Only HMM profiles are not provided by default because they might be very specific and should be given by the user instead. Thus, if you don't have any specific profiles that you want to screen for, just skip hmmsearch (which is done automatically by version 2.0.0).
For the pipeline test profile, we use the mybacteriocin HMM file from here which you could use as well, but it will screen only for mybacteriocins (as the name suggests), thus is not comprehensive regarding all AMPs.