Current fitting status and rough EDD of v1.2.0 release Slides: | @Hyesu Jang | Higher priorities Phosphate geometries are hard. The angle constants are probably too weak in FFs like GAFF. We should check those LPW + CIB -- Histograms might be more informative if we use a logarithmic y axis (to investigate tails of the distributions) DM – could show plot both with and without log scale on y axis
Later after May meeting CIB – Is there some way to make relative energies a”local” measure? Like, for some set of conformers in a molecule, we look at all of the inter-conformer relative energies. Then relative energies of conformers could be a sum of differences between that conformer and all conformers in the molecule. This would remove some o the sensitivity caused by choice of reference conformer. Note added by CIB: one could also include preferential weighting of lower-QM-energy conformers, a (good) implicit bias of the reference conformers we have been using so far. LPW + DM – That would be a good comparison to run after the May meeting.
|
