search

openforcefield / openff-toolkit

The Open Forcefield Toolkit provides implementations of the SMIRNOFF format, parameterization engine, and other tools. Documentation available at http://open-forcefield-toolkit.readthedocs.io
http://openforcefield.org
MIT License
311 stars 90 forks source link

Mulliken method frequently fails in antechamber #129

Closed j-wags closed 5 years ago

j-wags commented 5 years ago

A preliminary test of putting the molecule test database through antechamber using different charge methods indicated that the "mul" charge method failed on the ninth molecule it was given (ZINC00335972).

This can be replicated in test_molecule's test_compute_partial_charges function.

mkgilson commented 5 years ago

Do we need to bother with Mulliken? It is not a very useful method.

Mike

On 11/1/2018 10:59 AM, j-wags wrote:

A preliminary test of putting the molecule test database through antechamber using different charge methods indicated that the "mul" charge method failed on the ninth molecule it was given (ZINC00335972).

This can be replicated in test_molecule's test_compute_partial_charges function.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/openforcefield/openforcefield/issues/129, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AQEJQG2tzyS3kY1MkdT-HVDEJXVyPNFsks5uqzaJgaJpZM4YH3H5.

-- Michael K. Gilson, M.D., Ph.D. Professor and Chair in Computer-Aided Drug Design Co-Director UC San Diego Center for Drug Discovery Innovation Skaggs School of Pharmacy and Pharmaceutical Sciences 9500 Gilman Drive MC0736 La Jolla, CA, 92093-0736 Tel: 858-822-0622 gilson.ucsd.edu cddi.ucsd.edu bindingdb.org

j-wags commented 5 years ago

Probably not. I'm going to flag this as low-priority.

davidlmobley commented 5 years ago

Right, we don't actually even want to use any of the other charging methods in antechamber I think.

davidlmobley commented 5 years ago

(So an alternate fix to this and #128 would be to just raise an exception (NotImplementedError or some such?) if someone tries to use these and tell them that these are not supported and we recommend against using them.)