2020-11-12 WBO/Impropers meeting notes

Participants

• @David Mobley

• @Jessica Maat (Deactivated)

• @Trevor Gokey

• @Jeffrey Wagner

• @Pavan Behara

(Might be missing some points that were captured on the main page)

Forcefield refit for a single molecule, and iterate over all other molecules

• TIG0 can have one force constant

  • t43-45 have one k, and one periodicity

• (General) Do we have to worry about different torsion parameters that pass through the central bond?
  DLM: TIG0 is pretty general, we have wild cards on both ends so only this torsion parameter is relevant, and we do not have any other torsion parameters going through this central bond, don't have to worry about the extra complexity

• (General) Keeping all parsley constant and fitting only TIG0?
  DLM: Yes.

JW: Question on slack regarding mixing of interpolated with non-interpolated parameters

DLM: Applying an interpolated torsion multiple times may result in too deep or too shallow cases

How many times each param is being applied is also of interest.

Goal of TIG0 exercise: This smarts pattern being general captures all the points in subsequent TIG parameters (that have central CX3~CX3 bond). We can then make plots on these subsets since we already have the fitted torsions (or the k values). This information is more relevant than the current QM plots.
We may have to do the same exercise for the carbon-nitrogen central bond, as in place wildcards at the end and capture all points, then project them onto subsequent specific parameter plots.