Automated codon usage table selection

[mahesh-panchal]

Mitohifi and perhaps other tools need to specify the codon usage table to use.
Can this be automated?

[gbdias]

The simplest solution I could find is to use Entrez Direct efetch to search NCBI taxonomy using the taxonomy ID (that we currently get from the TAXONKIT_NAME2LINEAGE process:

efetch -db taxonomy -id ${taxid} -format xml | awk -F'[<>]' '/<MGCId>/ {print $3}' > mitocode.txt

nf-core has some of the Entrez Direct tools but not efetch. Entrez Direct is in bioconda so a container can be pulled from Seqera.

[mahesh-panchal]

Want to have a shot at writing either an nf-core module or a local module for it?

nf-core module

Make a fork of nf-core modules.

git checkout -b entrezdirect_efetch

Then make a new module

nf-core modules create entrezdirect/efetch

but it means you'll also have to write an nf-test too.

Locally

Make a fork of the EBP pilot workflow

git checkout -b entrezdirect_efetch
nf-core modules create entrezdirect/efetch

but here you won't have to write an nf-test and don't need to follow nf-core guidelines.

Then make PR back to the main workflow

[mahesh-panchal]

We can also query ENA:

# Get mitochondrial genetic code
SPECIES="Laetiporus sulphureus" # meta.id
curl -s "https://www.ebi.ac.uk/ena/taxonomy/rest/scientific-name/$( sed 's/ /%20/g' <<< "$SPECIES" )" | jq -r '.[] | .mitochondrialGeneticCode'

https://ena-docs.readthedocs.io/en/latest/retrieval/programmatic-access/taxon-api.html

and could be implemented similarly to https://github.com/NBISweden/Earth-Biogenome-Project-pilot/blob/main/modules/local/tol/search.nf