(205d) Simulation Studies of Pattern Recognition: What's the Problem?

Jayaraman, A. - Presenter, University of Colorado at Boulder

We are using computer simulation to study the thermodynamics of pattern recognition in biomimetic systems. Most of the theoretical work in the area of the pattern recognition by biomimetic polymers has focused on keeping the pattern on the heterogeneous surface fixed and allowing different types of copolymers to adsorb on the surface so as to determine those characteristics of a copolymer that can "recognize" the surface pattern. However, the question of how to design optimal surfaces or what pattern the surfaces should have for recognizing specific monomer sequences in copolymers had not been answered yet. To address this issue, we have developed a novel simulation method to design surfaces for recognizing specific monomer sequences in copolymers. The highlight of this work is that the designed surfaces recognize specific monomer sequences with higher selectivity than standard surfaces of sizes commensurate with the monomer sequence. This work provides the foundation for future work in understanding many biological processes that rely on pattern recognition, such as transmembrane signaling, pathogen-host interactions, viral-inhibition, etc.

Another system where molecular recognition plays a vital role is in DNA microarrays. DNA microarrays have been widely adopted by the medical research community for a variety of applications including identifying genes that are differentially expressed in healthy versus diseased cells, etc. In order to improve performance and to design next generation microarrays there is a need for a fundamental understanding of the interplay between the various factors that affect microarray performance. To gain such a understanding we study the thermodynamics and the kinetics of hybridization of single stranded "target" genes in solution with complementary immobilized "probe" DNA molecules on a microarray surface. Monte Carlo simulations on a coarse-grained lattice model are being used to examine how various parameters affect the extent of hybridization and the kinetics of the hybridization process. This work should give a fairly broad physical picture of molecular recognition in DNA microarrays and eventually provide a set of general guidelines for maximizing microarray sensitivity and specificity.