Improvement of the physical_validation package

Figure 1. The kinetic energy distribution obtained from a molecular dynamics trajectory and the corresponding analytical distribution given the simulation temperature. `physical_vadliation` provide functionalities for validating the kinetic energy, ensemble or the integrator of a molecular simulation. The figure here shows that the system indeed sampled the correct kinetic energy distribution.

Molecular simulations such as molecular dynamics (MD) and Monte Carlo (MC) simulations are powerful tools allowing the prediction of experimental observables in the study of systems such as proteins, membranes, and polymeric materials. The quality of predictions based on molecular simulations depend on the validity of the underlying physical assumptions. physical_validation allows users of molecular simulation programs to perform simple yet powerful tests of physical validity on their systems and setups. It can also be used by molecular simulation package developers to run representative test systems during development, increasing code correctness. The theoretical foundation of the physical validation tests were established by Merz & Shirts (2018), in which the physical_validation package was first mentioned. In this internal project within the Shirts Research Group, we improved the package by introducing more real-world examples in the documentation, tweaking the tutorials and enhancing the code coverage of the package.

Wei-Tse Hsu
Wei-Tse Hsu
Postdoctoral Research Associate in Drug Design

Computational Biophysicist keen on molecular dynamics, deep learning, and education