John Anthony Rose, PhD
The University of Tokyo, Faculty of Engineering,
Department of Computer Science and
Undergraduate Program in Bioinformatics and
7-3-1 Hongo, Bunkyo-ku,
Tokyo 113-0033, Japan
One of the primary research areas in Bioinformatics is the prediction of biopolymer native structure (the so-called ‘folding problem’). One of the most popular approaches for addressing this problem is the development of a statistical thermodynamic model, which combines a statistical picture of biopolymer interaction with experimentally measured energetic parameters, in order to assess the relative occupancies of the various conformations accessible to a biopolymer of interest. In my research, this approach is applied to model the hybridization of nucleic acid mixtures. Because the fields of DNA computing and nanotechnology provide a wealth of interesting and relatively well-defined nucleic acid mixtures for study, my attention has been focused on the modeling of DNA-computing-based systems and biotechnologies. My two systems of focus have been: (1) the DNA chip, especially Tag-Antitag Systems, and (2) Whiplash PCR, a simple autonomous DNA computer.
My recent published results include (please see Publications, for articles): (1) development of an equilibrium chemistry/statistical thermodynamic model for estimating the degree of undesired, or ‘error’ hybridization for the DNA Chip; (2) development of an evolutionary approach to apply this model to the solution of the associated ‘inverse folding problem’ for Tag-Antitag Systems (i.e., select DNA sequences which minimize error hybridization); (3) development of a theoretical model for assessing the impact of hairpin backhybridization on the efficiency of Whiplash PCR. In this model, statistical thermodynamics was applied to assess the probability of successful extension during each effective polymerase-DNA encounter, while the overall iterative extension process of each DNA hairpin was modeled as a Markov chain; (4) development and simulation of PWPCR, an enhanced-efficiency version of Whiplash PCR, which inhibits backhybridization by targeted PNA2/DNA triplex formation; (5) development and partial simulation of a PWPCR-based architecture for implementing an in vitro Genetic Program for solving instances of the NP-complete problem, Hamilonian Path; (6) development and partial simulation of a PWPCR-based Genetic Program for the in vitro evolution of proteins with a high affinity to a molecular target of interest, via constrained shuffling of protein ‘pseudomodules’. Current work is focused on theoretical refinement and simulation, as well as direct experimental validation of the predictions of these models.