Possibility of parameterizing crosslinkers

[aminsagar]

Hello.
Thanks for this excellent work.

I am trying to understand if there is a way to parameterize chemical crosslinkers with this approach. For example the one seen in 7UY2

I think one way would be to model the whole crosslinker bound to one of the cysteines as a single residue and then somehow add a bond to the second cysteine. Is there a way of doing this already present in the approach?

I would be really grateful for any suggestions.
Best,
Amin.

[j-wags]

Hi @aminsagar, thanks for trying the prototype out! I think there are a couple of ways to do this, but the approach of defining a new residue as cysteine+linker is the most direct way to go. You'll probably need two residue definitions:

• Cysteine+linker, with an extra crosslink bond defined from the linker to the second cysteine. If the crosslinking molecule follows standard PDB naming, you may be able to fetch its chemistry using its three-character residue name, and then use the react() method to combine it with the cysteine residue definition.
• A new CYS residue, which is identical to the original except for having a new crosslinking bond (instead of SG-SG, it'll be SG-(whatever the carbon in the linker is called))

A lot will also depend on whether the input PDB file has CONECT records for the atoms in the crosslinker. If so, you'll be able to get things working without needing to match the atom names in the PDB file exactly. But if not, then you'll need to make sure the new residue definitions do match the structure's atom names exactly.

There's more info on making custom residue definitions here. If you have a way to add hydrogens to your PDB file, feel free to share it on this issue and we can try to guide you further!