Molecular simulations have been widely applied to determine the structures and thermodynamic properties of organic compounds, the conformational characteristics and reactivity of proteins and nucleic acids, and the ligand-protein bindings. Still, the scope of the applications has been limited by the balance between accuracy and computational speed. In particular, the popular quantum-mechanical/molecular mechanical (QM/MM) approach has been largely limited to the single-configuration calculations due to the fact that QM computation is slow to be subject to dynamic sampling needed for the evaluation of free energy. In this SBIR application, we propose a software development based on a technology that will drastically shift this balance in QM/MM computation. Our new technology, called QM/MM-MFEP (minimum free energy path), will speeds up the combined QM/MM dynamic simulations hundreds of times. It achieves the efficiency gain by computing the dynamics of the MM subsystem separately from the QM subsystem such that the QM problem is solved with the averaged effect of the MM environment based on rigorous theory. In the Phase I of this project, we plan to make robust and efficient implementation of QM/MM-MFEP for solvation that will calculate accurately and efficiently the free-energy profiles of chemical reactions and relative free energies of ligands in solution, both of which are essential in the modeling of chemical and biological applications and practical applications in drug discovery. We will also improve the computation of the QM and MM electrostatic potential. In Phase II we will aim to enable free energy calculations of enzymatic reactions and ligand-protein binding affinities, and pKa's of ligands. The end product will be a QM/MM package that combines the best of the QM and MM, i.e. the QM accuracy and MM statistical simulation of the condensed phase, and afford its users routine simulations of reactions, binding affinities and solvation effects in condensed phases with high accuracy and significantly expand the reach of molecular modeling.
Public Health Relevance Statement: Public Health Relevance: We will develop a new molecular modeling technology that will dramatically improve the efficiency of the modeling. Molecular modeling is essential to the biological research and development and in drug discovery. The new technology will greatly enhance the quality and the productivity of the scientists in the practice of molecular modeling.
NIH Spending Category: Bioengineering
Project Terms: Affinity; anergy; Arts; base; Behavior; Binding (Molecular Function); Binding Proteins; Binding Sites; Biochemical Reaction; Biological; biological research; biological systems; Characteristics; Charge; Chemical Models; chemical reaction; Computer Simulation; Data; Data Compression; drug discovery; electron density; Electrostatics; Environment; Enzymes; Equilibrium; Evaluation; Fibrinogen; Free Energy; Hybrids; Hydrophobicity; improved; Lead; Ligand Binding; Ligands; Literature; Mechanics; Methods; methyl group; Modeling; Molecular; molecular dynamics; molecular modeling; Molecular Models; Nature; new technology; Nucleic Acids; Optics; Pharmaceutical Preparations; Phase; practical application; Process; Productivity; Property; Proteins; public health relevance; quantum; Reaction; Reading; Relative (related person); research and development; research study; Rest; Sampling; Scheme; Scientist; simulation; Small Business Innovation Research Grant; software development; Solutions; Solvents; Speed (motion); Staging; Structure; System; Technology; theories; Thermodynamics; Time; Trees
