PLACE:  171 Durham

SPEAKER:
Simon Tavaré
Departments of Biological Sciences,
Mathematics and Preventive Medicine,
University of Southern California

TITLE:
Seeing the Forest for the Trees

ABSTRACT:
 

There has recently been an unprecedented explosion in techniques for surveying molecular variation among organisms. These data allow us to study molecular processes on a wide variety of time scales. For example we can compare distantly related species, or individuals from the same species, or cells within a given individual. I will describe some statistical problems that arise from the study of such data, focusing in particular on ancestral inference what can we infer about molecular and historical processes using variation data? I will use several examples to illustrate the ideas, including reconciliation of fossil and molecular estimates of divergence times, estimation of the age of a mutation, and understanding the mutation mechanism in Huntington disease. A unifying theme is the study of highly dependent data in which the dependence is generated by a tree or graph.  Computational inference methods for dealing with the dependence will be described. This lecture is introductory; no specific knowledge of molecular biology will be assumed.

Reference:  Tavaré, S. (1999).  Random trees in  molecular genetics.  Bull. ISI., 52, 269-272.
 

COFFEE:  3:45 p.m., 104 Snedecor