Expanding the capabilities of the Molecular Simulation Design Framework (MoSDeF)

About the Project

The Molecular Simulation Design Framework (MoSDeF, mosdef.org) is an open-source, extensible Python-based software package designed to facilitate the hierarchical construction of both atomistic and coarse-grained system configurations, the application of classical force fields (i.e., atom-typing), encapsulation of chemical topology information, and the generation of input files for a variety of molecular dynamics and Monte Carlo simulation engines, including LAMMPS, GROMACS, HOOMD-Blue, Cassandra, and GOMC. The use of MoSDeF for developing and running molecular simulations enables the simulations to be reported in a reproducible way, which is becoming an increasingly important requirement for computational science. MoSDeF has many tools that serve the research community performing biological simulations, including methods for creating coarse-grained models of complex biomolecules and applying forcefields used in biosimulations.

This project involves developing new capabilities for MoSDeF inspired by the needs of biological simulations and integrating other open-source tools relevant to biosimulations. MoSDeF has an active worldwide developer community, so coordination with other developers is an important aspect of this project. Multi-developer software creation is done through the public repository GitHub. Students thus gain valuable experience in how to develop software using modern, industry-standard multi-developer tools.

Utilising MoSDeF and open-source simulation engines the student will:

• develop new capabilities for performing materials simulations with the MoSDeF tool suite
• examine ways to facilitate and perform multi-scale simulations that span the molecular to meso-scales with easy interoperability
• perform molecular simulations to test the tools and examine novel materials of interest, with an emphasis on soft materials (e.g., molecular liquids, polymers, and biological tissues that are easily deformed at relevant application conditions)

Applicants should have some experience in writing computer codes (such as Python or MATLAB). Applicants need not have prior molecular simulation experience, though familiarity with open-source codes such as GROMACS and HOOMD is an advantage.