Professor Andreas Hilfinger
University of Toronto, Mississauga
Friday, February 22, 2019 - 11:00am
Ramsay Wright Building, Room 432
Many cellular processes are complex yet only partially characterized. Constructing realistic models thus often requires making a large number of assumptions based on speculation. Instead of ignoring or guessing unknown details, I will show how to rigorously characterize classes of models that share some features but are left to vary arbitrarily in all unknown features. I will demonstrate how such an approach can be used to non-perturbatively test hypotheses about biochemical rates in complex processes against observed cell-to-cell variability. For example, we analyzed gene expression variability in E. coli and showed observed protein level fluctuations cannot be explained by varying levels of cognate mRNA. The second application of such an approach is to establish hard performance bounds that limit the ability of cellular processes to control cell-to-cell variability. I will illustrate how such trade-offs give us insights into the design principles of cellular processes even when our knowledge of them is necessarily incomplete.
Professor Vince Tropepe
Dept of Cell and Systems Biology