Boston University | Center for Computational Science
HomeNews and EventsResearchEducationPeopleSeminarsFacilitiesContact Us

Energy, Errors and Dynamics in Computation by DNA Self-assembly

Erik Winfree
California Institute of Technology
February 18, 2000
Thermal fluctuations and diffusion are major sources of noise in molecular systems, and any molecular system that performs reliable computation must have mechanisms for counteracting this noise. We examine this issue in an abstract model of DNA self-assembly, wherein universal computation can be performed by the sequence-directed self-assembly of DNA into a 2D crystal. The basic stochastic steps in our model are based on the kinetics of DNA hybridization. We find that temperature and DNA concentration are the critical parameters determining the specificity of binding reactions, and thus the success or failure of the computation. Furthermore, we can evaluate the error rate and speed of self-assembly, as well as the energetic cost per computational step. Interestingly, the energetic cost per step approaches zero near the melting temperature of the crystal. We hope to compare these predictions to laboratory experiments in the near future.

copyright © 2006, Center for Computational Science | Boston University , MA, 02215