-
Notifications
You must be signed in to change notification settings - Fork 23
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Unclear origin of 90- and 270-degree angle torsion phases #53
Comments
This is odd; I will have to think about this more, but it's not obvious to me that we should ever have 90/270 degree phase angles (unless we have both at the same time?!?)... would be interested if @pavankum has any insight. I would have also asked Chris Bayly but it seems he was in the other thread and has this question also...? |
Unfortunately the forensics look pretty tricky, at least for me as an individual only digging through these files after the fact. Starting with what @j-wags noticed happened with
The changelog for 1.0.7 includes only:
Footnotes
|
For the record, we have ongoing conversations about this. After discussion in a meeting last week, we more or less concluded that:
So this is in the works. These don't seem to be a "bug" per se but more of a "legacy issue we've inherited" and probably most shoud be removed until/unless we have data indicating they are needed. |
From openforcefield/openff-toolkit#1370 (comment), it is not clear why some torsions use 90 or 270 degree phases. It is not intuitive that these parameters would have "handedness" and thus far we have failed to understand or justify the added complexity. We should aim to document somewhere why these parameters do not use the typical 0/180 degree phases; we currently risk losing the reasoning here to the sands of time.
The toolkit may or may not have different behavior when using
label_molecules
versus actually applying parameters, but that is a separate topic from why these parameters have handedness at all.The text was updated successfully, but these errors were encountered: