Abstract

Quantum mechanical (QM) calculation is now an important tool for investigations of functional mechanisms of biological macromolecules based on their three dimensional and electronic structures. However, the system size which QM calculations can treat is usually up to a few hundred atoms, whereas those of most biological systems of interests are in the range of 1,000 to 1,000,000 including surrounding solvent water. To overcome these difficulties, quantum mechanics/molecular mechanics (QM/MM) calculation has been used as an efficient method, in which the system is divided into QM and MM regions; the active sites to be investigated are assigned as the QM regions, which are described quantum mechanically, and the other regions of the macromolecular systems are assigned as the MM regions, which are described molecular mechanically. To date, many works for developments of efficient/accurate algorithms and their implementations/applications have been performed for QM/MM calculations. In this study, we have developed an interface program to connect conventional but highly-parallelized QM and MM calculation engines running on massively-parallel supercomputers with more than thousands of CPUs. We connected AMBER and GAMESS for molecular dynamics (MD) and QM calculation engines, respectively, which enabled us to perform high-performance QM/MM hybrid MD simulations. Actually, we have evaluated the accuracy and performance of the present system on our supercomputers, the PACS-CS system (14.34 TFlops) and the T2K Tsukuba system (95.39 TFlops) in the Center for Computational Sciences, University of Tsukuba, by comparing the calculated results with experimental data with respect to a metalloprotein (the Cu-bound active center is assigned as the QM region). Furthermore, we have applied it for investigations of environmental effects on the electronic structure of a protein-DNA complex and the reaction mechanisms of cytochrome c oxidase.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call