W4400y - Biological Networks
The course will start with an introduction to types of biological networks and many of the new high throughput and quantitative technologies now available.
We will start with the mathematical and computational analysis of small networks in order to understand some of the basic principles in biological networks including network motifs, modularity, robustness and stochasticity.
The course will then scale up to much larger networks teaching the computation techniques needed to address these including Hidden Markov Models, Bayesian networks, FDR, Bootstrapping, Expectation Maximization, Inference, Gibbs Sampling, Monte Carlo and Belief Propagation. We cover many of the pitfalls of high throughput data and how to over come these, proper modeling choices when building large scale models of molecular networks and how to apply the techniques learned to real data.
We will learn how to reconstruct regulatory networks from such data and understand how these networks compute, dynamically change and the connections between genetic sequence and these molecular regulatory networks. Finally will demonstrate how the Bayesian techniques learned in the course can be applied to other biological networks such as a network of interacting neurons.
Number of Credits: 4
Semester: spring 2010
Section Number: 001
Location: 337 Mudd
Days of the Week: Tuesday & Thursday
Start Time: 1:10 pm
End Time: 2:25 pm
Course Website: https://courseworks.columbia.edu/cms/outview/courseenter.cfm?no=BIOLW4400_001_2010_1
Instructor: Dana Pe'er