GENESIS: a hybrid‐parallel and multi‐scale molecular dynamics simulator with enhanced sampling algorithms for biomolecular and cellular simulations

<jats:p><jats:styled-content style="fixed-case">GENESIS</jats:styled-content>(Generalized‐Ensemble Simulation System) is a new software package for molecular dynamics (<jats:styled-content style="fixed-case">MD</jats:styled-content>) simulations of macromolecules. It has two<jats:styled-content style="fixed-case">MD</jats:styled-content>simulators, called<jats:styled-content style="fixed-case">ATDYN</jats:styled-content>and<jats:styled-content style="fixed-case">SPDYN</jats:styled-content>.<jats:styled-content style="fixed-case">ATDYN</jats:styled-content>is parallelized based on an atomic decomposition algorithm for the simulations of all‐atom force‐field models as well as coarse‐grained Go‐like models.<jats:styled-content style="fixed-case">SPDYN</jats:styled-content>is highly parallelized based on a domain decomposition scheme, allowing large‐scale<jats:styled-content style="fixed-case">MD</jats:styled-content>simulations on supercomputers. Hybrid schemes combining<jats:styled-content style="fixed-case">OpenMP</jats:styled-content>and<jats:styled-content style="fixed-case">MPI</jats:styled-content>are used in both simulators to target modern multicore computer architectures. Key advantages of<jats:styled-content style="fixed-case">GENESIS</jats:styled-content>are (1) the highly parallel performance of<jats:styled-content style="fixed-case">SPDYN</jats:styled-content>for very large biological systems consisting of more than one million atoms and (2) the availability of various<jats:styled-content style="fixed-case">REMD</jats:styled-content>algorithms (T‐<jats:styled-content style="fixed-case">REMD</jats:styled-content>,<jats:styled-content style="fixed-case">REUS</jats:styled-content>, multi‐dimensional<jats:styled-content style="fixed-case">REMD</jats:styled-content>for both all‐atom and Go‐like models under the<jats:styled-content style="fixed-case">NVT</jats:styled-content>,<jats:styled-content style="fixed-case">NPT</jats:styled-content>,<jats:styled-content style="fixed-case">NPAT</jats:styled-content>, and<jats:styled-content style="fixed-case">NPγT</jats:styled-content>ensembles). The former is achieved by a combination of the midpoint cell method and the efficient three‐dimensional Fast Fourier Transform algorithm, where the domain decomposition space is shared in real‐space and reciprocal‐space calculations. Other features in<jats:styled-content style="fixed-case">SPDYN</jats:styled-content>, such as avoiding concurrent memory access, reducing communication times, and usage of parallel input/output files, also contribute to the performance. We show the<jats:styled-content style="fixed-case">REMD</jats:styled-content>simulation results of a mixed (<jats:styled-content style="fixed-case">POPC</jats:styled-content>/<jats:styled-content style="fixed-case">DMPC</jats:styled-content>) lipid bilayer as a real application using<jats:styled-content style="fixed-case">GENESIS</jats:styled-content>.<jats:styled-content style="fixed-case">GENESIS</jats:styled-content>is released as free software under the<jats:styled-content style="fixed-case">GPLv2</jats:styled-content>licence and can be easily modified for the development of new algorithms and molecular models.<jats:italic>WIREs Comput Mol Sci</jats:italic>2015, 5:310–323. doi: 10.1002/wcms.1220</jats:p><jats:p>This article is categorized under:<jats:list list-type="explicit-label"><jats:list-item><jats:p>Structure and Mechanism > Computational Biochemistry and Biophysics</jats:p></jats:list-item><jats:list-item><jats:p>Computer and Information Science > Computer Algorithms and Programming</jats:p></jats:list-item><jats:list-item><jats:p>Molecular and Statistical Mechanics > Molecular Dynamics and Monte-Carlo Methods</jats:p></jats:list-item></jats:list></jats:p>