|
Fall 2008
STATISTICS
COLLOQUIUM
Wednesday, September 3, 2008
3:30-4:00—Refreshments
4:00-5:00—Talk
Yost Hall, Room 101
Tomas Radivoyevitch, Ph.D.
Professor, Department of Epidemiology and Biostatistics
Case Western Reserve University
Equilibrium Model Selection
Ribonucleotide reductase (RNR) is precisely controlled to meet the dNTP demands of scheduled (replication driven) and unscheduled (repair driven) DNA synthesis. It has a small subunit R2 (45 kDa) that exists almost exclusively as a dimer, and a large subunit R1 (90 kDa) that dimerizes when dTTP, dGTP, dATP, or ATP binds to its specificity site, and hexamerizes when dATP or ATP binds to its activity site. In general, RNR is modeled as a pre-equilibrium of proteins, ligands, and substrates whose parameters of interest are dissociation constants Kd, and a set of turnover rate parameters kj that map distributions of active enzyme complexes into expected kcat measurements of mixtures. Because the masses of R1 and R2 are known, it is logical to focus first on Kd estimation from protein oligomer mass measurements, and later on kj estimation from enzyme activity measurements. 58 a priori plausible equilibrium models of dTTP-induced
R1 dimerization were derived from a full model of two total concentration constraint quadratic equations in the two free concentrations [dTTP] and [R1]. This talk will describe this model space, the methods I used to fit it to available data, and how the top
6 models suggest subsequent experiments to discriminate between them.
|
