Current Research Projects

DNA Melting and Hybridization

The recent development of DNA sensor microarrays has the potential to significantly benefit important applications, such as mutation detection, RNA expression and genetic sequencing. DNA sensors are composed of single-stranded DNA oligomers bound to a surface. The development of this technology relies on knowledge of DNA denaturation or ``melting'' and hybridization transitions, and their sensitivity to many variables has left several questions unanswered, such as the mechanism by which changes in physical environment between solution and bound DNA act to influence hybridization, or competitive adsorption of nearly identical sequences. DNA melting is a phenomenon that occurs when hybridized double-stranded DNA helices dissociate into single-stranded DNA coils. The thermodynamic melting temperature is defined as the temperature at which the double-strand is exactly half-dissociated and can be measured quite accurately experimentally. However, the mechanisms and structures governing the melting process are less well understood. The application of a simplified model with one interaction site to represent the base and another site for the phosphate + sugar backbone allows the investigation of melting and rehybridization transitions. The tendency for strands to misalign is also captured with this approach in a natural way. Monte Carlo simulations using configurational bias Monte Carlo (CBMC) and self-adapting fixed-endpoint CBMC allow the efficient sampling of phase space for these systems.

The image is of two complementary 10-nucleotide oligomers from work by Nick Tito.

Tunable Supercritical Fluid Separation Systems

Most chemical processes involve solvents and the high cost of separating reactants, products, catalyst and solvents and the negative environmental costs of today's solvents are motivation to develop tunable and environmentally benign solvent systems and chemical processes. These processes attempt to use fairly innocuous materials such as supercritical carbon dioxide and polyethylene glycol to achieve what traditional processes do with more detrimental materials. Unfortunately, the solvating power of supercritical fluids is generally lower than that of liquid solvents. Through laborious experiments it has been found that their performance can be improved. For example, it has been observed that small amounts of cosolvents or ``entrainers'' can greatly enhance the solvating power and selectivity of supercritical solvents, but it is difficult to determine their effect without actually carrying out the experiment. Monte Carlo molecular simulations allow the efficient modeling of such systems with the goal of optimizing solvation conditions. This is done by carrying out calculations with varying thermodynamic (e.g. temperature and/or pressure) conditions and compositions.

Shown here is carbon dioxide (thick bonds) solvation of ethyl benzene (black spheres) in poly(ethylene glycol) (red and blue spheres) within a 9 A radius around ethyl benzene.