Software

With support from NSF at Vanderbilt we have been developing the Molecular Simulation Design Framework (MoSDeF),  a Python-based framework designed to make it easier to construct and initialize soft matter systems, vary structural parameterization to enable screening, and capture all inputs and processes such that a simulation can be considered TRUE (Transparent, Reproducible, Usable by others and Extensible). More information about the TRUE concept can be found in a recent open access paper here.

MoSDeF is built around several distinct tools that provide functionality for the construction of arbitrary soft matter configurations (mBuild) and application of force fields (Foyer), which can integrates with the Signac toolkit developed by the Glotzer group, allowing screening over a user-defined parameter space). 

By providing researchers with systematic methods and libraries to initialize their simulation systems, MoSDeF makes this step more scriptable, automatable, and hence, reproducible. The guiding principles that underlie the development are encoded in the TRUE acronym that is promoted by the project: to produce molecular simulation results that are Transparent, Reproducible, Usable by others, and Extensible. By hosting the open and version-controlled source code on GitHub (https://github.com/mosdef-hub), we are encouraging users to interact and contribute to the project in a transparent manner.

MoSDeF is now under development by research groups at 7 institutions through an NSCI Framework award titled “Software for Building a Community-Based Molecular Modeling Capability Around the Molecular Simulation Design Framework (MoSDeF).”