News

Introduction

Biologically important phenomena, such as complex formation and conformational change, occur on a time-scale of milliseconds to seconds. Therefore, it is difficult to reproduce these phenomena in conventional molecular dynamics simulations. Multicopy and multiscale molecular dynamics simulation methods enable it on the basis of statistical mechanics. μ2lib has been developed to facilitate the development of novel algorithms of multicopy and multiscale molecular dynamics simulation methods and the implementation of these algorithms.

The initial virsion of μ2lib (mu2lib-K-1.0) was developed at Molecular Scale Team, Computational Science Research Program, RIKEN. The source code is distributed under the GNU General Public License at Riken ISLiM Web site.

After the program ended in 2013, the μ2lib developers' team continues its development. The current members of the developers' team are:

The source code of the latest version of μ2lib is distributed through this site.

Features

Because existing MD programs are designed to perform a simulation on only one copy of a system, we developed a new software from scratch. Considering that the multicopy and multiscale methods are still developing, we developed a versatile library, instead of developing a monolithic program.

μ2lib is written in C++. It is a set of classes, which are classified into core, interface, and application class groups.

Core class group

This group is composed of:

  • Classes describing data structures of force-field parameters, molecular topologies and coordinates, and simulation settings
  • Classes for calculating forces and energies of standard interatomic interactions and restraints
  • Classes describing specific algorithms such as the SHAKE, genelarized Born, and smooth particle mesh Ewald methods
  • Classes for conducting energy minimization and various types of molecular dynamics simulation
  • Classes for intra- and inter-copy communication

Interface class group

The classes of the core class group read and write files in original formats. For the convenience of users, μ2lib provides interfaces for reading and writing files in other formats. The following file formats are supported.

  • Amber parameter/topology file format
  • Amber configuration file format
  • Amber restart file format (text)
  • Amber trajectory file format (text/netCDF)
  • Gromacs trajectory file format (xtc/trr)

All the classes in this group are subclasses of the core classes. Therefore, the user can easily add new classes that support other file formats.

Application class group

μ2lib has already been used to develop several application programs, including the programs of the multiscale enhanced sampling (MSES) method and on-the-fly string method (see Publications page). This group contains the classes used in the application programs. Most of them are also subclasses of the core classes.

μ2lib is distributed under the GNU General Public License. Go to Download page to obtain the source code.

Acknowledgment

The development of μ2lib was partly supported by Research and Development of the Next-Generation Integrated Simulation of Living Matter, by Strategic Programs for Innovative Research Field 1 “Supercomputational Life Science”, and by JSPS KAKENHI. The developers thank Dr. Hiroyuki Sato at Information and Mathematical Science and Bioinformatics Co., Ltd. for his technical support.

Feedback

The μ2lib developers' team welcomes feedback about μ2lib. Please send questions, comments, and bug reports to