chrisiacovella
commented
6 years ago I think we should also add some text somewhere in our readme for foyer that says some variant of the following things:
1- if you are taking parameters from a force field file (e.g., distributed with a simulation engine), you should validate the parameters by examining the original source of the parameters (i.e., the journal article that presented the parameters), as force field file may contain errors.
2- Note the original source may utilize slightly different functional forms (e.g., whether or not there is a 1/2 in front of harmonic terms). The conventions used in our format are: … 3- Note, the original source may have inaccurate labels. In particular units for harmonic angle constants are often mislabeled (e.g., labeled as kcal/mol/degree^2 when in fact the parameters correspond to kcal/mol/rad^2). Spring constants for harmonic angles are typically on the order of 100 kcal/mol/rad^2. Please take note of any inconsistencies that are found (via comments, desc, and notes associated with the force field you are implementing). Our format follows openMM and uses the following units: ...
OpenMM has the functional forms here: http://docs.openmm.org/latest/userguide/theory.html#standard-forces We should also include the functional forms somewhere in our documentation since the Foyer XML is a super set of the OpenMM fields with more extensive metadata + smarts