Problem with processing amide bonds for backbone N-methylated residues

[ackbar03]

Hi,

I am working with some residues that have the backbone nitrogen atom N-Methylated.

I noticed there seems to be some problems when using Boltz1 where the amide bond from the modified residue isn't processed correctly. It doesn't seem to be an issue in other models such as AF3 or Chai1 however.

---

For example, when looking at N-METHYLCYSTEINE (CCD NCY: https://www.rcsb.org/ligand/NCY), it seems to work fine in Chai1.

It also seems to work fine in AF3 (the bonding is slightly strange because PyMOL determines bonds based on atomic distance, but the atomic configuration is correct)

However, the amide bond formed by the oxygen doesn't seem to be properly processed in Boltz

---

We can also see in Chai 1, N-METHYLPHENYLALANINE (CCD MEA: https://www.rcsb.org/ligand/MEA) produces something that more or less makes sense. (The drawn bonds are a bit messy)

It is also correct in AF3

In Boltz1 however, the amide bond formed by the Oxygens on the modified are again not processed.

---

This looks to me like a post-processing issue that should be fixable. Is it possible the team take a look and see what the problem is and if its possible to provide a fix? Thanks for your work!

[ackbar03]

Hi,

I saw there was a recent fix regarding the OXT atom in #93. This fixed the problem.

Thanks