Methods are presented for simulating chemical reaction networks with a spatial resolution that is accurate to nearly the size scale of individual molecules. Using an intuitive picture of chemical reaction systems, each molecule is treated as a point-like particle that diffuses freely in three-dimensional space. When a pair of reactive molecules collide, such as an enzyme and its substrate, a reaction occurs and the simulated reactants are replaced by products. Achieving accurate bimolecular reaction kinetics is surprisingly difficult, requiring a careful consideration of reaction processes that are often overlooked. This includes whether the rate of a reaction is at steady-state and the probability that multiple reaction products collide with each other to yield a back reaction. Inputs to the simulation are experimental reaction rates, diffusion coefficients and the simulation time step. From these are calculated the simulation parameters, including the 'binding radius' and the 'unbinding radius', where the former defines the separation for a molecular collision and the latter is the initial separation between a pair of reaction products. Analytic solutions are presented for some simulation parameters while others are calculated using look-up tables. Capabilities of these methods are demonstrated with simulations of a simple bimolecular reaction and the Lotka-Volterra system.

Vibrational Stark effects, which are the effects of electric fields on vibrational spectra, were measured for the C−N stretch mode of several small nitriles. Samples included unconjugated and conjugated nitriles, and mono- and dinitriles. They were immobilized in frozen 2-methyl-tetrahydrofuran glass and analyzed in externally applied electric fields using an FTIR; details of the methodology are presented. Difference dipole moments, Δ μ, equivalent to the linear Stark tuning rate, range from 0.01/f to 0.04/f Debye (0.2/f to 0.7/f cm-1/(MV/cm)) for most samples, with aromatic compounds toward the high end and symmetric dinitriles toward the low end (the local field correction factor, f, is expected to be similar for all these samples). Most quadratic Stark effects are small and negative, while transition polarizabilities are positive and have a significant effect on Stark line shapes. For aromatic nitriles, transition dipoles and Δμ values correlate with Hammett numbers. Symmetric dinitrile Δμ values decrease monotonically with increasing conjugation of the connecting bridge, likely due to improved mechanical coupling and, to a lesser extent, an increased population of inversion symmetric conformations. Δμ values are close to those expected from bond anharmonicity and ab initio predictions.

Most cellular processes depend on intracellular locations and random collisions of individual protein molecules. To model these processes, we developed algorithms to simulate the diffusion, membrane interactions, and reactions of individual molecules, and implemented these in the Smoldyn program. Compared to the popular MCell and ChemCell simulators, we found that Smoldyn was in many cases more accurate, more computationally efficient, and easier to use. Using Smoldyn, we modeled pheromone response system signaling among yeast cells of opposite mating type. This model showed that secreted Bar1 protease might help a cell identify the fittest mating partner by sharpening the pheromone concentration gradient. This model involved about 200,000 protein molecules, about 7000 cubic microns of volume, and about 75 minutes of simulated time; it took about 10 hours to run. Over the next several years, as faster computers become available, Smoldyn will allow researchers to model and explore systems the size of entire bacterial and smaller eukaryotic cells.