Rare Event Sampling and Related Topics III

November, 12-13 2015

The Institute of Statistical Mathematics

Tachikawa, Tokyo


Seminar Room 5 (third floor)

Yukito Iba (ISM)
Hiroshi Fujisaki (Nippon Medical School, ISM)

Abstract of Invited Lecture

Luca Maragliano (Istituto Italiano di Tecnologia)

TALK 1: Methods for potential of mean force and reaction paths calculations in molecular dynamics simulations

Molecular Dynamics Simulations are a powerful approach to study biomolecular processes. Straightforward applications, however, are often hampered by incomplete sampling, since in a typical simulated trajectory the system will spend most of its time trapped by high energy barriers in restricted regions of the configuration space, making only rare transitions between them. Over the years, several simulative techniques have been designed to overcome this problem and enhance space sampling. In this talk, I will review a set of approaches designed to characterize rare events in complex systems via the exploration and reconstruction of their free energy landscape, or of transition paths on them, with a focus on the case of numerous collective variables. I will outline the methods and their implementation and illustrate applications to ligand diffusion and conformational changes in biomolecules.

TALK 2: Molecular Dynamics Simulations of synaptic proteins and their role in neuronal signaling

Neuronal signal transmission occurring at synapses is a process orchestrated by a complex molecular machinery. Signaling arises from release of neurotransmitter carried by synaptic vesicles, which undergo a well-tuned trafficking cycle. After being filled with transmitter, the vesicles dock and fuse to the cell membrane to allow its release. These steps are then followed by different routes of recycling that prepare the vesicles for the subsequent round. The entire cycle is controlled by the action of a large number of synaptic proteins, as for example the members of the SNARE family. In this talk, I will review recent progress made in the study of vesicle cycling proteins, underlying the contribution provided by Molecular Dynamics Simulations. In particular, I will focus on the role of Synapsins, a family of proteins involved at different stages of the trafficking cycle.

Dr. Luca Maragliano's web site

November 12 (Tur.)

Yukito Iba (ISM)

Luca Maragliano (Istituto Italiano di Tecnologia)
TALK 1: Methods for potential of mean force and reaction paths calculations in molecular dynamics simulations


Yukito Iba (ISM)
Rare event sampling using multicanonical MCMC and time-reversed simulation

November 13 (Fri.)

Luca Maragliano (Istituto Italiano di Tecnologia)
TALK 2: Molecular dynamics simulations of synaptic proteins and their role in neuronal signaling

(lunch break)

Nobu C. Shirai (Mie University)
Lattice gas model for -synuclein fibril formation under crowded environment


Tohru Terada (Univ. of Tokyo)
The protein-ligand binding process studied by coarse-grained molecular dynamics simulation and the string method

Kei Moritsugu (Yokohama City Univ.)
Energy landscape of protein-ligand Interaction revealed by multiscale enhanced sampling

Aim and Scope

The art of realizing rare events in computer simulation and the calculation of their occurrence probabilities are important in many fields of science and engineering. The goal of this series of symposiums is to provide a coherent perspective on this subject, focusing on novel computational techniques and their applications to highly complex and nonlinear systems.

Recently, there has been interest in rare event sampling in the field of molecular simulation. Typical examples are numerical studies on the transition paths of conformational changes of a protein molecule. In this symposium, a keynote lecture by Dr. Luca Maragliano (Istituto Italiano di Tecnologia), as well as several talks by leading Japanese researchers in this field has been scheduled.

On the other hand, there are a number of other application fields of rare event sampling. Specifically, recent introduction and innovative use of Markov Chain Monte Carlo (MCMC) promotes a variety of novel applications in these fields. Some interesting examples are found in studies on chaotic dynamical systems, random graphs, random matrices, and counting combinatorial objects, as well as other hard computing problems in information theory, machine learning, meteorology, and astronomy. In this symposium, we also investigate these subjects and algorithms that have been developed for them.

Data assimilation and the design of computer experiment are subjects closely related to rare event sampling. In a nutshell, data assimilation selects a set of simulation sample paths that are compatible to observed data from the real world, whereas rare event sampling selects a set of sample paths that are compatible to target rare events. In the adaptive design of computer simulations, a most informative set of initial conditions and parameters are selected using past simulation results. We also include these subjects, aiming at cross-fertilization among these topics and rare event sampling.

iba at ism.ac.jp