User:Sooihk/sandbox

William L. Jorgensen

Research Interest

Molecular modeling of organic and biomolecular systems using BOSS and MCPRO

William Jorgensen was pivotal in the development of computational programs such as BOSS (Biochemical and Organic Simulation System) and MCPRO (Monte Carlo for proteins). These programs orginate from a Monte Carlo software written by Jorgensen for the simulation of pure liquids in the 1970s.

BOSS is a molecular modeling system capable of performing various calculations, such as enegry minimizations, normal mode analysis, and conformational searching. The BOSS utilizes Monte Carlo (MC) statistical mechanics simulations, and mixed quantum and molecular mechanics (QM/MM) calculations. QM/MM calculations implement force fields and statistical mechanics to model systems that contain thousands of atoms with QM ability to describe inherent interactions and bond breaking and forming events. This is achieved by dividing the chemical system into two regions. Reacting solutes or monomers are described by quantum mechanics. The surrounding solvent molecules and nonreactive solute molecules are described by the molecular mechanic OPLS-AA force field. Metropolis MC statistical mechanics is then used to statistically sample the whole chemical system. These hybrid calculations overcome conventional separate QM and MM limitations. Hybrid calculations allow large chemical and biological systems to be modeled whereas they would be too expensive to model by QM alone. Integrating QM with MM or MC allows the QM region to fluctuate with the environment which is useful when investigating solvent effects on organic reactions.

MCPRO is a derivative of BOSS and retains most of its basic functionalities. However, MCPRO specializes in efficient Monte Carlo simulations of biomolecules in solutions and makes use of the concept of residues. The Monte Carlo simulations can be performed in a solvent box or cluster. Energy minimizations are performed with conjugate gradient with analytical derivatives. The Monte Carlo simulations can be executed for pure liquids, solutions, clusters, or gas-phase systems. Applications include computing the structure and thermodynamic properties of a liquid at a specified temperature and pressure, free energies of solvation, effects of solvation on relative energies of conformers, locations of transition structures for reactions in solution, changes in free energies of solvation along reaction paths, and structures and relative free energies of binding for host–guest complexes in solution. Monte Carlo methods rely on repeated random sampling to obtain results. The Monte Carlo sampling in BOSS and MCPRO use standard procedures including the Metropolis algorithm. For a periodic system of flexible molecules, the Monte Carlo method randomly picks one molecule, translates it randomly in all three Cartesian directions, randomly rotating the molecule about one randomly chosen axis, and rebuilding the molecule from its Z-matrix with random modifications to any bond lengths, bond angles and dihedral angles.

Impact on Field of Science and Scientific Community(writing a part of it)

Programs like BOSS and MCPRO are heavily utilized in computational chemistry and drug discovery. Hybrid quantum mechanics and molecular mechanics (QM/MM) computer simulations have become a vital tool for studying chemical and biological phenomena for systems too large to treat with QM alone. Programs like BOSS and MCPRO tether independent QM and MM software packages which allow ab initio and density function methods into QM/MM simulations. BOSS and MCPRO have been used in every field of chemistry to describe molecular interactions. The ability to carry out QM/MM has allowed researchers to model large chemical and biological systems that would have otherwise been to expensive to model by QM alone.