Chapin Cavender

icon indicating a website link https://openforcefield.org/

icon company Open Force Field Initiative icon location Postdoctoral fellow with Open Force Field Initiative and Gilson lab at UC San Diego. Molecular dynamics, computational chemistry, and biopolymer force fields.

Results 6 comments of Chapin Cavender

Carboxylates from mol2 files produced by tleap have incorrect bond orders and formal charges

> in the future we could have `[C-]([O-])([O-])` motifs lead to an informative error. I think this is the best way to address this in the long run. Throw a...

bespokefit gives the torsion a worse fit than the OpenFF

I think this result is expected based on how BespokeFit weights grid points in the objective function via ForceBalance. Briefly, grid points are weighted according to the QM energy with...

Make HMR simulations easier

For the GROMACS engine in the `proteinbenchmark` repo, @ajfriedman22 developed [a workflow](https://github.com/openforcefield/proteinbenchmark/blob/70e55c8311c0b862cc6fde3634d3f49f85d5bb26/proteinbenchmark/system_setup.py#L772-L783) that exports to OpenMM and then uses the `parmed` python API to apply HMR and export to GROMACS....

Chemically equivalent atoms with delocalized formal charge are assigned different torsions

Following up on some suggestions from Chris Bayly, it looks like phosphate esters and sulfate esters are assigned correctly because there is only one torsion SMIRKS that matches them. The...

Benchmark proteins (eventually other biomolecules) using F@H

I'll provide some more detail about the anticipated simulation needs. Based on previous protein force field benchmarks, we will estimate NMR observables (chemical shifts, scalar couplings, and NOEs) from unbiased...

Benchmark proteins (eventually other biomolecules) using F@H

My preference for this use case is to have full trajectories as output and do the analysis locally. You can use multiple forward models for each type of observable, and...