ColabFold icon indicating copy to clipboard operation
ColabFold copied to clipboard

complex prediction with individual MSAs per protein

Open SAMtoBAM opened this issue 7 months ago • 1 comments

Hi there

I am trying to query a set of proteins for complexes against a larger proteome database I have precomputed MSAs (a3m) files for each of the individual proteins How can I provide them to the colabfold_batch step when providing the complex as the fasta input (proteins seperated by ':')

Thank you!

SAMtoBAM avatar Sep 16 '25 15:09 SAMtoBAM

I got a similar problem, except I managed to generate each a3m file with the same a3m protein complexed with different smaller proteins/peptides (one per file), but models are not running the predictions for the complexes. For example I got: Running prediction for complex: D3VML5p Using custom MSA file: /content/drive/MyDrive/AMP/D3VML5p.a3m 2025-09-25 17:54:40,948 Running on GPU Error running prediction for complex D3VML5p: not enough values to unpack (expected 4, got 2) Any ideas to map the issue will be most welcome. THANKS

juliocoll avatar Oct 03 '25 17:10 juliocoll