Tel-Aviv University - Computer Science Colloquium

Monday, Jan 9, 2006, 16:15-17:15

Schreiber 309


Ziv Bar Joseph

Carnegie Mellon University


Data integration for understanding dynamic systems in the cell



Dynamic systems, such as the cell cycle and immune response, play an

important role in many biological processes. Recent advances in

high-throughput experimental methods are enabling researchers to obtain a

global view of the temporal expression profiles of such systems. Using time

series expression data we were able to model some of these dynamic systems

in yeast.  However, when moving from model organisms to humans we face many

new computational challenges. Human systems are more complex, their temporal

duration is longer and the data is often noisier. Using ideas from machine

learning and graphical models I will present algorithms that combine time

series data with additional datasets, which are often static, to address

issues ranging from experiment design to data analysis to pattern

recognition and modeling.