It's possible to do that, though it may not be faster, especially if you're running on a GPU. The cost of memory access versus computation is much higher on a GPU than on a typical CPU. But you can try it and see how it works. To do that, you precompute the combined parameters for each pair of atom ...

CustomNonbondedForce can't do PME. It supports arbitrary functional forms. PME can only be applied to a very small class of specific functional forms, specifically inverse integer powers of distance. Presumably you want to use the Buckingham potential in place of Lennard-Jones, while still using a s...

Read https://github.com/openmm/openmm/wiki/Frequently-Asked-Questions#template, which explains what's going on. There are two distinct things here: the definition of the molecules to simulate, which comes from the PDB file, and the definition of the force field, which comes from the XML file. The tw...

Setting parameters to 0 is not the same as an exception. Exceptions are used for 1-2, 1-3, and 1-4 interactions. They tell the nonbonded kernel to skip calculating an interaction. Setting the parameters for a particle to 0 doesn't cause anything to be skipped. It just guarantees that all interaction...

It still has the same problem described above. Your PDB file labels the residue as DG, claiming that it's a standard residue. But of course it isn't. OpenMM tries to add the standard bonds expected in a DG residue, but there are no bonds connecting the S atom.

When your system includes multiple nonbonded forces (NonbondedForce, CustomNonbondedForce, CustomGBForce), they need to exclude exactly the same interactions. That's because they all get computed together in a single GPU kernel based on a single neighbor list. If one of them excludes an interaction,...

Sure, you can post your code here.

I don't know if any of the openmmtools developers monitor this forum. It might be best to open an issue at https://github.com/choderalab/openmmtools.

I see several problems with your PDB file: - It lists all the atoms with HETATM rather than ATOM. C and G are standard residues, but the file claims they're heterogens. - The hydrogens are named incorrectly. They just have arbitrary numbers like H01 and H02. - The molecule isn't terminated correctly...

If the LJ force weren't being applied, the distance between opposite charges would shrink without limit until the massive forces caused the simulation to explode. You set sigma to 0.227 nm for Cl and 0.141 nm for Na, so distances in the range of 0.2 to 0.3 nm are about what you expect. What were you...