Mating by the numbers: quantitative measurements and computational modeling of the pheromone response.
Kirsten R. Benjamin (1), Larry Lok (1), Ty Thomson (2), Drew Endy (2), Roger Brent (1)
(1) Alpha Project, Molecular Sciences Institute, 2168 Shattuck Ave., Berkeley, CA, 94704, USA;
(2) Division of Biological Engineering & Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139
We combine experimental and computational methods to deepen our understanding of the response to mating pheromone in S. cerevisiae. One ultimate aim is to predict the behavior (or output) of a cell in response to a defined perturbation (or input) like pheromone treatment or mutation. Detailed prediction requires plugging a quantitative model into a computational simulation of the dynamics of pathway events. The model is built by enumerating events in a genetic pathway as a series of linked chemical reactions (e.g., binding, unbinding, localization, catalysis) and specifying the quantities involved (number of molecules, reaction rates). To constrain the models, we have carefully measured the abundances (molecules per cell) of 15 key proteins in the pathway. We are using two different methods to simulate pathway dynamics: (1) a continuous method that consists of solving a series of ordinary differential equations (ODEs) that describe pathway events and (2) a discrete stochastic event simulator, called Moleculizer, that keeps track of the thousands of individual complexes formed from pathway proteins and generates a series of intermolecular events by Monte Carlo methods. We will present our initial answers to two questions: How does pathway behavior change in response to changes in the levels of pathway components? How do changes in the representation of the pathway (ODE, Moleculizer) impact models of pathway behavior?.
Return to YGM 2004 Home at SGD