Reproducibility

While every effort has been made to ensure the 'correctness and reproducibility' of any results computed using this framework, achieving consistent free energies between different frameworks and simulation engines has been notoriously tricky ¹.

In an attempt to ensure that this framework remains at least self-consistent between versions, and as consistent as possible with other packages, a suite of regression tests are provided in the main GitHub repository.

These include tests to ensure that computing the free energies using both the 'equilibrium' and 'non-equilibrium' methods supported in this framework are in agreement amongst themselves, and with the GROMACS values reported by Loeffler et al ¹.

Regression Results

The results of running the free energy comparisons using the latest version of the framework are shown below:

The average hydration free energies of methane, methanol, toluene, and 2-cyclopentanylindole computed using the 'equilibrium' and 'non-equilibrium' methods across three replicas. The error bars show the standard deviation computed across three replicas.

---

1. Hannes H Loeffler, Stefano Bosisio, Guilherme Duarte Ramos Matos, Donghyuk Suh, Benoit Roux, David L Mobley, and Julien Michel. Reproducibility of free energy calculations across different molecular simulation software packages. Journal of chemical theory and computation, 14(11):5567–5582, 2018. ↩↩