EPSRC grant for rare event simulations

An illustration of the Forward Flux Sampling method.
An illustration of the Forward Flux Sampling method.

A team of researchers from the Institute of Condensed Matter and Complex Systems and EPCC has been awarded £297,000 by EPSRC to develop software for simulating rare events in soft matter and biological systems.

The grant was awarded to Kevin Stratford, Rosalind Allen and Davide Marenduzzo, who will work with postdoc Juho Lintuvuori on the 2-year project.

Rare events are processes that happen unpredictably and infrequently, but may be very important when they do happen. Examples include computer system failures, earthquakes and disease outbreaks. Despite their importance, rare events are notoriously difficult to study in computer simulations, because of the long waiting times between events. A method developed by Rosalind Allen and coworkers, called Forward Flux Sampling, alleviates this problem by allowing simulations to sample only the interesting trajectories where the event happens, and not the waiting times.

This EPSRC High Performance Computing grant will allow the team to implement the Forward Flux Sampling method in two widely used simulation packages: the Edinburgh-developed Ludwig code for Lattice-Boltzmann simulations and the LAMMPS Molecular dynamics simulation code, in order to study transitions between different liquid crystal phases, and the nucleation of chiral crystals under shear. Eventually the researchers hope to develop a version of the method that is portable to a wide variety of different simulation packages, opening the door to applications across condensed matter and biological physics.