Transition Path Sampling Simulations via PyRETIS

Theory and Application of Rare Events Methods to Compute Transition and Reaction Rates

11 - 15 March 2019

Venue: Lorentz Center@Oort

If you are invited or already registered for this workshop, you have received login details by email.

We offer a workshop on rare events in molecular simulations. The theoretical key concepts, with a special regard to Transition Path Sampling and its latest developments (e.g. Replica Exchange Transition Interface Sampling),  will be discussed [1]. A newly developed free and open-source software, PyRETIS (http://www.PyRETIS.org) will be presented, its usage explained, its application assisted.

            In chemical physics, a rare event is a transition between two long-lived states (i.e. highly meta-stables or stable states) that happens very infrequently due to energetic and entropic bottlenecks. Complementary limitations in experimental, theoretical, and computational approaches still hamper a direct and quantitative exploration of all important steps that occur in a rare event. Molecular dynamics simulations, which in principle could provide insights at high resolution, can only achieve timescales up to ~102- 103 ns for reasonably large systems (>105 particles). An efficient computational description of such complex processes can be achieved by focusing on the transition dynamics between the states. Transition Path Sampling methods does not require a preconceived notion of either the physical-chemical mechanisms or of the transition states, nor the inclusion of artificial forces/potentials. Information on reaction mechanism, transition states, rate constants and reaction barriers of the process of interest can be therefore obtained without alter the natural dynamics of the systems.

            Since its conception, several new algorithms based on the original path sampling method have been developed [2]. Among these, the Replica Exchange Transition Interface Sampling (RETIS) [3],  is particularly efficient. The main strength of RETIS is that it is exact: the estimated rate converges to the one obtainable by long and straightforward MD simulations, but orders of magnitude faster.

            We propose to organize an interactive school to promote the usage of the RETIS algorithm through the software PyRETIS developed by the group of Titus van Erp (http://www.van-erp.org) at NTNU in Trondheim. PyRETIS is a free and open-source software package (http://www.PyRETIS.org) and has already been interfaced to a few simulation engines (GROMACS, CP2K) performing, therefore, simulation both via classical MD or Ab Initio MD (further external simulation engines will be soon also supported). Successfully applications will be presented, such as silicate oligomerization [4], water autoprotonization [5], DNA-protein binding.

[1] Cabriolu R., Skjelbred R. K. , Bolhuis P.G. , and van Erp.  T. S., J. Chem. Phys., 147, 15, 2017

[2] Chandler D., Classical and quantum dynamics in condensed phase simulations, 1st Chapter, 1998.

[3] van Erp T., Moroni D. and Bolhuis P. G., J. Chem. Phys. 118, 7762, 2003.

[4] M. Moqadam, E. Riccardi, T. T. Trinh, A. Lervik, and T. S. van Erp, J. Phys. Chem. Chem. Phys., 19, 13361, 2017

[5] Moqadam M, Lervik, A, Riccardi E, Venkatraman V, Alsberg K, and an Erp.  T. S., Proc. Natl. Acad. Sci., 115, 20, E4569, 2018.

Read more...


Follow us on:

Niels Bohrweg 1 & 2

2333 CA Leiden

The Netherlands

+31 71 527 5400