CDS prediction but not annotation option about bakta

[oschwengers]

Hi @PedroMTQ ,
I see your point here but you could simply take the CDS protein or nucleotide sequences from the .faa/.ffn files and use them with any external annotation tool of your choice. Might that be a way to go?
Of course you could also just use Prodigal to predict CDS sequences, however in this case you would miss the sORF sequences extracted by Bakta.

[PedroMTQ]

Hi @oschwengers ,
Yes I could do that, but isn't Bakta also doing CDS functional annotation? I'd just like to skip that step (just to minimize runtime) while preserving all other Bakta's features. If implementing such a feature is inconvenient, I could of course do as you said.

Thanks and regards,
Pedro

[oschwengers]

sure, Bakta uses Prodigal for the regional prediction and then conducts several subsequent steps for the functional annotation (sequence identification and homology searches with subsequent lookup of pre-assigned annotations as well as expert annotation systems).
Implementing another skipping option should be possible. However, I'd like to keep the usage as small/simple as possible and I'm not yet quite sure if this is a larger use case. Therefore, I'll leave this open just in case that more users might ask for such a feature - otherwise I would kindly ask to use the CDS sequences as suggested above.

Thanks for reaching out!