1
$\begingroup$

Are there established practices for fine-tuning AMBER MD parameters in scenarios like mine, involving specific factors like mutant proteins or non-standard ligands? Has anyone successfully tackled challenges in a similar protein-ligand system?

Additionally, I'd appreciate insights on recommended approaches for addressing specific challenges and optimizing the simulation setup. The ultimate goal is to ensure that MD simulations precisely capture the nuances of protein-ligand interactions while maintaining computational efficiency.

$\endgroup$

1 Answer 1

2
$\begingroup$

If you're working with Amber, have a look at some of the tutorials, for example:

which should walk you through setting up a simulation with both non-standard amino acids and with arbitrary ligands.

In addition to GAFF/GAFF2, you might want to try OpenFF for ligands, using openff-toolkit (the parameters are compatible with Amber for the protein).

About some of the specific things you ask:

"Are there established practices for fine-tuning AMBER MD parameters" - typically (outside of modified residues) the protein forcefields are used as is, they're not fine-tuned

"while maintaining computational efficiency" - if not using a polarizable forcefield (rare) or four-site water model (also rare), any set of parameters runs at about the same speed; there isn't an accuracy/efficiency tradeoff , since essentially all the common forcefields (Amber, CHARMM, GAFF, OpenFF, CGenFF) have the same nonbonded functional form, and most of the time is spent on computing nonbonded interactions

"Has anyone successfully tackled challenges in a similar protein-ligand system?" - of course, it is a huge and active field of research

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .