April 1, 2008
EE Conference Room (MUDD 1306)
Speaker: Michael Samoilov
Biological networks have a number of unique and distinguishing properties, which encompass a wide variety of structural, functional, and control features that have been evolutionarily selected to robustly support life under a broad spectrum of conditions. Yet, underlying this overt phenotypical diversity are elementary chemical reactions and physical interactions that link together the assorted molecular species and complexes with which biological systems are composed. Remarkably, it can be shown that this basic characterization leads to a number of powerful biochemical and biophysical constraints that profoundly restrict the range of possible biological network designs compatible with corresponding regulatory, dynamic and other empirically observed traits. The resulting modeling framework provides novel understanding of engineering principles behind many natural biological systems and offers new tools for identifying their organizational features from phenotypical information. Applications of this approach range from improving reverse-engineering of gene regulatory pathways, to the analysis of signal processing capabilities of ubiquitous biochemical motifs, to investigation of molecular virulence mechanisms in prominent pathogens. Examples include biological networks/circuits from E. coli, B. subtilis, S. cerevisiae, Drosophila and HIV, among others.
Michael Samoilov is a Research Staff Member at the California Institute for Quantitative Biosciences (QB3) at UC Berkeley. Dr. Samoilov earned his Bachelor’s degree with Honor in Physics and Mathematics from Caltech (1991). He then went on to do graduate work at Stanford University, beginning with high-energy physics and astrophysics, for which he was awarded an M.S. in Physics (1994), and continuing at the biophysics program where he received a Ph.D. in Biophysics (1997). After spending several years developing stochastic trading strategies for leading finance companies and running a Webby Award-winning multimedia start-up, Michael was drawn back to science by the emergence of novel biological systems engineering and analysis paradigms driven by the advances in single-molecule, single-cell and bulk high-throughput experimental methods. His most recent work includes investigating the role of discrete and stochastic dynamics in biological circuits, developing biochemically - and biophysically-driven methods for structural identification and functional analysis of biological networks, as well as studying information and signal processing characteristics of biomolecular reaction systems.