You are right that the ChimeraX AlphaFold PAE plotting requires the pairwise score matrix to have rows for each non-polymer atoms. So ChimeraX won't currently plot pairwise scores for Model Archive structures with non-polymer residues. This could be fixed. But as you note the most important case to fix will be for AlphaFold 3 PAE values and currently ModelCIF does not have a table that can handle the mix of per-residue and per-atom PAE values that AlphaFold 3 reports. So let's wait until ModelCIF figures out how to handle AlphaFold 3 PAE scores then we can fix ChimeraX reading those from ModelCIF. I made a ChimeraX ticket for this issue

On Oct 11, 2024, at 9:30 AM, Gerardo Tauriello <gerardo.tauriello@unibas.ch> wrote:

Dear Tom,

Checking out the daily build, I noticed that files without local-pairwise values do not have a clean error handling anymore.
E.g. "open ma-asfv-asfvg-001 from modelarchive pae true" is an example of a file without local-pairwise values and fails at "metric_id = met_ids[0] if met_ids else values[0][5]" because met_ids is empty in that case...

Also there is a caveat for files like ma-tbvar3d-15 which include a non-polymer. There I think that "alphafold pae" expects per-atom PAE values for the non-polymer and so we end up with a PAE JSON file with a wrongly sized matrix. That one is an edge case within ModelArchive and so it can be considered low priority until we properly handle AF3 PAE matrices in ModelCIF.

Regards,
Gerardo