Unclear fasta spec

[jkosinski]

AlphaPulldown/README.md

Line 52 in 8b465a3

- **FASTA file paths** instead of UniProt IDs: `/path/to/protein.fasta`

It is a bit unclear whether the fasta file can have multiple proteins/sequences or it should be one fasta per protein. I don't know and could you rephrase to make it clearer (but still concise)?

Related to that, if multi-protein fasta is not supported yet by the snakemake pipeline, this should be reported as an issue to fix.

[DimaMolod]

No, fasta files with multiple entries can't be used with snakemake by definition (1 file - 1 job). For that exact reason @maurerv added s splitter script to the repo:
https://github.com/KosinskiLab/AlphaPulldownSnakemake/blob/main/config/split_fasta.py

[jkosinski]

I think we still need a solution so that users do not need to split manually. Otherwise, it adds an extra complexity. Would it be possible to have the splitting as an extra optional workflow step? It would be conceptually equivalent to retrieving sequences from uniprot.