Molecular modeling and simulation of large systems remains a very difficult task because of complexity and diversity of biological macromolecules. Despite of ever increasing computational capability, investigations at the atomic level pose many unresolved issues. Currently, classical force fields represent most popular approach to studying of biomolecular systems. However, molecular mechanics (MM) force fields are unable to describe the changes in the electronic structure of a system undergoing a chemical reaction. The changes in molecular topology or intermolecular charge transfer require quantum mechanics (QM) for the proper treatment. However, due to high computational cost, the application of ab initio QM is still limited to relatively small systems. Correspondingly, combined QM/MM methods have attracted a significant attention in the past. However, there are apparent limitations on this path as well. Difficulties in formulation of the QM-MM interaction scheme, bond breaking while defining the QM zone, size of the QM zone, and consistent treatment of solvation bring certain difficulties to the field. Therefore, significant efforts have being made to develop new schemes that would allow treating the entire system at the approximate quantum-mechanical level. 

We present a linear scaling semiempirical method LocalSCF [1],[2] and the derivation of the underlying variational finite localized molecular orbital (VFL) approximation [3]. Due to economical use of computational resources LocalSCF method made possible QM calculation of 100,000 atoms systems on a desktop computer. Based on the VFL approximation, and particularly on its ability to treat each linear coefficient of molecular orbital independently, we developed the novel type of the semiempirical QM/QM method in which a part of the system, including ligand and protein active site, are treated at the full self-consistent regimen while the protein bulk is considered as carrying a frozen electronic density matrix. The developed QM/QM method is implemented in the LocalSCF suit of programs [4] and applied toward the protein-ligand binding energy prediction searching through the library of 20,000 ligands represented in total by 200,000 conformations [3]. This remarkable computational performance makes feasible routine application of the QM theory to the computational studies of large biological systems and structure based drug design.

[1] Anikin, N.A.; Anisimov, V.M.; Bugaenko, V.L.; Bobrikov, V.V.; Andreyev A.M.; J. Chem. Phys. 121 (2004) 1266.

[2] Anisimov, V.M.; Panczakiewicz, A.; The Linear Scaling Semiempirical LocalSCF Method and the Variational Finite LMO Approximation, Challenges and Advances in Computational Chemistry and Physics, Vol. 13 (submitted).

[3] Anisimov, V.M.; Bugaenko, V.L.; J. Comp. Chem. 2009, 30, 784.

[4] Bugaenko, V.L.; Bobrikov V.V; Andreyev A.M.; Anikin, N.A.; Anisimov, V.M.; LocalSCF, ver 2.1, Fujitsu Limited, 2007, Tokyo, Japan.

• Legal notice:
  

  Copyright (c) Pielaszek Research, all rights reserved.
  The above materials, including auxiliary resources, are subject to Publisher's copyright and the Author(s) intellectual rights. Without limiting Author(s) rights under respective Copyright Transfer Agreement, no part of the above documents may be reproduced without the express written permission of Pielaszek Research, the Publisher. Express permission from the Author(s) is required to use the above materials for academic purposes, such as lectures or scientific presentations.
  In every case, proper references including Author(s) name(s) and URL of this webpage: https://science24.com/paper/23146 must be provided.