Dr. Albert S. Benight

A major goal of modern medicine is the design of multiplexed nucleic acid diagnostic assays for identification of specific genetic features that are associated with particular (diseased or abnormal) phenotypic traits. The outstanding molecular recognition capabilities of DNA inherent in the specific sequence of base pair interactions that endow its genetic function DNA also dictate its utility and performance in hybridization-based assays. Nucleic acid diagnostic assays based on multiplex hybridization have the potential to revolutionize genomic research and genetic medicine. Multiplex assays offer never before imagined capabilities for systematic high throughput screening, discrimination and analysis of large numbers of DNA (and RNA) sequences. Although multiplex hybridization will form the basis of nearly all modern nucleic acid diagnostic assays, there is currently a substantial gap in our understanding of the underlying reaction chemistry and physics of multiplex mixtures of DNA containing more than two single strands.

Intrinsically, multiplex hybridization reactions are more complex than the more standard "simplex" hybridization reaction scheme where only two perfectly complementary strands are present that anneal with one another to form a perfectly matched duplex; or anneal with each other to form mismatched dimers. These increased complexities associated with multiplexing arise mainly because, when given the opportunity, DNA is promiscuous in binding with other strands. In a multiplex reaction, the sheer numbers of different strands present provides overwhelming opportunities for mismatched duplex formation or cross-hybridization. For instance, if there are N different probe and N perfect match target sequences in the reaction mixture, the number of possible different mismatch cross-hybrids is at least N² - N. As a result, both the kinetic and thermodynamic behaviors of individual hybridization events are grossly affected by the potential competition that can occur between perfect match and mismatch strands in the multiplex hybridization environment.

Another major focus of our research program is on the development of quantitative descriptions of multiplex hybridization reactions that consider the entire "system" behavior in a collective manner. In our approach multiplex hybridization is considered to be a competitive, multi-channel, reaction process: a system wherein many species can react both specifically and non-specifically with one another. Among other important features, the analysis provides critical assessments of errors associated with cross-hybridization and their consequences on results. These quantitative analytical treatments have been applied to design and analyze experimental results of multiplex hybridization reactions performed on microarrays; and clearly demonstrated reactions between each probe:target do not occur in isolation as separate uncoupled events.

General analytical and robust numerical solutions have been developed that support both equilibrium and kinetic models of multiplex hybridization systems comprised, in principle, of any number of targets and probes. These modern and realistic models of multiplex hybridization facilitate enhanced design, analysis and optimization of superior multiplex, nucleic acid diagnostic assays.

The DNA diagnostics and microarray industries are rapidly emerging and projected to generate revenues in the tens of billions of dollars in just a few years. Before the broad utility and enormous commercial potential of microarray-based assays can be fully realized, microarrays must be elevated from their current status primarily as hypothesis generators; to the level of quantitative discovery tools. Many of the hurdles and challenges currently facing microarray diagnostics are within the current scope of our research program. Our overarching aim is to make microarray diagnostics quantitative thereby enabling microarrays to be widely used discovery tools for high throughput genomic screening, genotyping, forensic analysis; and pharmacogenomic applications.