A Search-Based Approach for Bayesian Inference of the T -cell - PowerPoint PPT Presentation

BIOT -2007 A Search-Based Approach for Bayesian Inference of the T -cell Signaling Network Bradley Broom Mitchell Koch Dept. of Bioinf. & Comp. Biology Dept. of Computer Science M.D. Anderson Cancer Center Rice University bmbroom@mdanderson.org mkoch@rice.edu Devika Subramanian Dept. of Computer Science Rice University devika@rice.edu

Signaling & metabolic networks • Consist of interacting proteins, genes and, small molecules • Underlie the major functions of living cells • Goal : learn these networks from experimental data, particularly how they are altered in diseased cells

Challenges • The cell is a complex stochastic domain: signal transduction, metabolic and regulatory pathways all interconnected. • We only observe mRNA levels and/or protein levels. • Measurements are noisy. • Limited amount of data.

Building models from data A HIGH HIGH … B LOW MED … C HIGH LOW … B A C

Bayesian network P(B HI ) = 0.8 B B P(A HI ) A HIGH 0.9 A B P(C HI ) LOW 0.2 HIGH HIGH 0.99 HIGH LOW 0.9 C LOW HIGH 0.5 LOW LOW 0.1

Learning Bayesian networks • Scoring function • Impossible to find highest scoring network • Super-exponential number of possible networks • Use heuristic search procedures

Identifying Significant Edges • Very many high-scoring networks • Each over-fitted to data • Find edges that occur frequently in good networks • Two approaches: • Markov Chain Monte Carlo (MCMC) • Bootstrap aggregation • Select edges occurring above threshold t

A Search-Based Approach for Bayesian Inference of the T -cell Signaling Network