CSCE Seminar: PNA-Mediated Whiplash PCR:
Massively Parallel Autonomous Molecular Computation
John A. Rose Institute of Physics
The University of Tokyo 3:30 PM Friday
April 27 Rm. 304 ENGR

Computers implemented in DNA have the potential for massively
parallel computational operations and vast amounts of storage capacity. In the
Whiplash PCR architecture, finite automata are implemented in the form of template-directed,
recursive polymerase extension of a mixture of DNA molecular hairpins, appropriately
encoded with finite state transitions. Autonomous parallel computation on a
massive scale is achieved by iterating the reaction temperature through a 3-step
thermal cycle appropriate for enforcing the parallel hybridization, extension,
and denaturation of all hairpin species. Although the autonomous nature of the
WPCR architecture shows great promise from a computational point of view, a
serious barrier which confronts the efficient implementation of various forms
of recursive-PCR-based computation is a systematic tendency for computational
molecules towards a simple form of self-inhibition. In this talk, modifications
to the protocol based on PNA are discussed that might make massively parallel
autonomous computation achievable in practice. In addition, an evolutionary
algorithm based on WPCR is proposed.