Host: R. Perez-Jimenez

Atomistic molecular dynamics (MD) simulations provide a uniquely detailed tool for understanding the dynamics of proteins, the workhorses of living organisms. Analyzing the results from these simulations is however an overwhelming task, as it involves making sense of many gigabytes of data consisting of coordinates of the many degrees of freedom of both the protein and the surrounding solvent. Markov state models have recently emerged as an analysis methodology for MD simulations that is able to provide information of the slow, and usually most relevant, transitions of the system (e.g. folding), but without loosing the resolution on the microscopic dynamics. One of their advantages is their fine-grained resolution, which allows for exquisite comparison of the simulation results with experiment.

I will introduce this approach using a very small peptide as example¹. Studying this system we have been able to calculate the rate of the most fundamental process in protein folding, helix nucleation, and understand the origin of the experimental fluorescent signal. Then we will gradually move up in complexity to show how these models can help us to setup the simulations for maximum efficiency in rate calculations² and to solve fundamental questions in protein folding, in particular the origin of the unusual viscosity dependence observed experimentally³. Finally I will present applications of this type of approach to protein engineering in systems of industrial interest^4,5.

 

1.    D De Sancho & R B Best, What is the time scale for ?-helix nucleation? J Am Chem Soc 133, 6809-6816 (2011).

2.    D De Sancho, J Mittal & R B Best, Folding kinetics and unfolded state dynamics of the GB1 hairpin from molecular simulations. J Chem Theory Comput 9, 1743-1753 (2013).

3.    D De Sancho*, A Sirur & R B Best, Molecular origin of "internal friction" effects in protein folding rates. Nat Commun 5, 4307 (2014).

4.    A. Kubas, D. De Sancho, R. B. Best & J. Blumberger, Aerobic damage of [FeFe] hydrogenases: activation barriers for O2 chemical attachment. Angew Chem Int Ed 53, 4081-4084 (2014).

5.    D De Sancho, A Kubas, J Blumberger & R B Best, Identification of hot-spots for mutation for substrate diffusion in proteins (Submitted).