Comparing Interaction Energy Analysis: QM vs. Force Field Approaches

User discussions
, , , ,
1

How can one perform an interaction energy (IE) analysis using GROMACS to compare with QM data? For example, consider the interaction energy curve of a cation with a molecule like methanol. QM calculations are conducted in a vacuum at 0 K, where the interaction energy (IE) is computed as: IE = E(dimer) - E(fragment1) - E(fragment2).

For the force-field part, one could emulate a vacuum calculation by using a very large simulation box and large cutoffs. Then, vary the cation-ligand distance in a 0-step molecular dynamics run (mdrun) for each distance step.

Are these results comparable? How so? How does one calculate the interaction energy using the force-field approach?

I’ve seen this type of curves in e.g. https://pubs.acs.org/doi/epdf/10.1021/acs.jpcb.2c07237.

QM-MM-IC
QM-MM-IC1510×748 120 KB

2

In general, gmx energy does this. It’s what this Gromologist workflow uses under the hood to make things simpler:

Whether they are comparable in practice - they should be, but it might depend on what the target of the original parameterization was. Some models are fine-tuned to reproduce energetics in vacuo, some to reproduce free energy profiles in water (or some combination of these two), and some to reproduce average molecular properties.