(253bd) A Logic-Based Framework for Defining Force Field Usage Semantics and Atom-Typing Molecular Systems

The availability of forcefields for molecular simulation has reduced the effort researchers must devote to the difficult and costly task of determining the interactions between species, allowing them to instead focus on the motivating scientific questions. However, determining which parameters in a force field to use is still often a tedious and error prone task. This difficulty is related to the strong dependence of the parameters on the chemical context of the atoms, where the chemical context may depend on the local bonded environment of an atom in a molecule, the type of molecule(s) being considered, the phase of the molecule(s), etc. While tools for performing atom-typing exist, they are often tied to a specific force field or simulation software and typically employ a rigid hierarchy of encoded â??if/elseâ? statements where rules that identify more specialized atom types must be called in precise order, such that more general atom types are only chosen when more specialized matches do not exist. Rather than relying on rigid rule hierarchies that are difficult to maintain, debug, and evolve, we instead present a general, force field agnostic toolâ??Foyerâ??that relies upon logic-based annotations of chemical context. In Foyer, rigid rule hierarchies are replaced by an iterative process which allows rules to be more flexibly embedded within the actual force field files, alongside the parameter definitions, thus delivering annotations that are both human and machine readable. This also allow rules to be tested for logical consistency, independent of the chemical species they actually describe. Here, we present several case studies that demonstrate the forcefield annotation scheme and general software usage.