Mesoscopic-level Simulation of Dynamics and Interactions of Biological Molecules Using Monte Carlo Simulation

Autor: Yamaguchi, Yoshiki, Maruyama, Tsutomu, Azuma, Ryuzo, Yasunaga, Moritoshi, Konagaya, Akihiko
Zdroj: The Journal of VLSI Signal Processing; 20240101, Issue: Preprints p1-13, 13p
Abstrakt: Abstract: A mesoscopic-level method for clarifying living cell dynamics is described that uses Monte Carlo simulation of biological molecule interactions. The molecules are described as particles that take a random walk in 3-dimensional discrete space. Many kinds of molecules (including complex forms) are supported, so complex reactions with enzymes can be simulated. Also described is an field programmable gate array system with reconfigurable hardware that that will support complete modeling of an entire cell. Two-phase processing (migration and reaction) is used to simulate the complex reactions, so the method can be implemented in a limited amount of hardware. The migration and reaction circuits are deeply pipelined, resulting in high performance. Estimated performance is 30 times faster than with a 3.2-GHz Pentium 4 computer. This approach should make it possible to eventually simulate cell interactions involving one billion particles.
Databáze: Supplemental Index