Alternate protonation states cause sequence truncation

[faraguna-ginkgo]

I have taken P42212 and modified the PDB file to introduce alternate protonation states:

• residue 90 to GLH
• residue 48 to CYX
• residue 139 to HIE

The fasta file from LigandMPNN, run with --redesigned_residues A48 A90 A139 includes sequences that are shorter by an amino acid, resulting from the removal of the E90. This includes both the reference sequence in the second line as well as all generated mutants.

I would suggest checking for protonation states and erroring out explicitly, or casting them to the standard residues.

ligand-out.zip