BCAM Alumni & Former Members Seminar: Markov State Models for understanding long-timescale dynamics of molecular systems

Host: Prof. Elena Akhmatskaya

Title
Markov State Models for understanding long-timescale dynamics of molecular systems

Abstract
Markov State Models (MSM) are a powerful framework for analyzing dynamical systems, such as molecular dynamics (MD) simulations. MSMs can reproduce the long-time statistical conformational dynamics of biomolecules using data from MD simulations that are individually much shorter. MSMs can predict both stationary and kinetic quantities on long timescales, and facilitate the extraction of insight into biomolecular mechanisms such as binding or folding. A MSM represents a master equation framework, meaning that using just the MSM the entire dynamics of the system can be described. The MSM itself is an n × n transition probability matrix where the entire configuration space spanned by the system has been divided into n states. By determining the states, one can track the dynamical progress of a system by writing down which state it occupies at time points separated by a lag time τ . For the lag time τ to be Markovian, the system must be memoryless: the probability that, after the next increment of τ , the system transitions to state y given it is in state x cannot depend on where the system was before it entered state x.

In this talk, we show recently developed applications of MSMs to the large and functionally important class of proteins called Intrinsically Disordered Proteins (IDP). IDPs lack a fixed or ordered three-dimensional structure and often lead to the aggregation of misfolded proteins that could result in diseases. Amyloid beta (Aβ) is a popular IDP due to its presence in amyloid plaques found in the brains of people with Alzheimer’s disease. Aβ aggregates irreversibly to form fibrils that compose such plaques. On the other hand, it is well known that there exist low complexity domains involved in liquid-liquid phase separation which form functional and reversible aggregates. In our case, we focus our attention to Low sequence complexity Aromatic-Rich Kinked Segments (LARKS).

Our aim is to understand better the two aggregation mechanisms and what are the structural differences that make the amyloid beta and the LARKS aggregate in different matters. Thus, by means of MSMs, we compare the binding mechanisms and, ultimately, the fibril formation of Aβ and the LARKS. We show promising results that both explain the physical process that we are interested in and show the potentiality of the MSMs. The presented results are part of a joint collaboration with Xabier López and David De Sancho (DIPC and UPV/EHU).