How does MetaProdigal handle metagenomics where different genetic codes are used?

I’m wondering how prodigal when using -p meta can handle genes from different genetic codes? For example, if Gracilibacteria/SR1 was in there it would use genetic code 25. Would it catch this? When running in meta mode does it end up using the genetic code translation table argument?

My understanding: Prokka will by default use translation table 11. It tries to detect codon usage and may switch to table 4 automatically. You should see a warning if the tool cannot make sense of the input with either. As far as i know there is no autodetection for other tables. You can use the argument -g 25 to specify table 25 manually. I don’t know a prodigal-only solution for when your Gracilibacteria are part of a metagenome, though. You would have to do some taxonomic binning first (which is recommended before prodigal usage anyways).