Statistics 215: Statistical Models in Biology 

Fall 2010   ●   T Th 11:00-12:15    ●   Sequoia Hall Room 200

Nancy R. Zhang   ●  userid: nzhang domain name: stanford.edu ●   Office Hours:  Tues  2:15-3:15 PM Sequoia Hall 141   


Course content:

Markov chains in discrete and continuous time, branching processes, and Poisson processes. Applications to models of nucleotide evolution, the Wright-Fisher process, coalescence, epidemiology, and sequence analysis. Theoretical material approximately the same as in STATS 217, but emphasis is on examples drawn from applications in biology, especially genetics.

Announcements:

 

Announcements will be posted here.

 

Teaching Assistant:

 

We have no TA for this class.

 

Required Text:

 

Hoel, Port and Stone, (1972) Introduction to Stochastic Processes.  Houghton Mifflin Company.

 

References:

 

Taylor, H.M. and Karlin, S. (1998) An Introduction to Stochastic Modeling, Academic Press: San Diego.

 

Tentative Plan: 

 

Week

Tentative topics (HPS = Hoel, Port and Stone)

 

 

 

 

1

Tues

Course overview,  Introductory examples (HPS 1.1) 

 

 

Thurs

Discrete Time Markov Chains: HPS 1:  Basic definitions, Chapman-Kolmogorov Equations. 

 

2

Tues

Discrete Time Markov Chains: HPS 1.5-1.6:  Recurrent and Transient States.

 

 

Thurs

Discrete Time Markov Chains: HPS 1.7-1.8: Extinction in birth and death and branching chains.

 

3

Tues

Discrete Time Markov Chains: HPS 2: Stationary Distributions. 

 

 

Thurs

Discrete Time Markov Chains:  HPS 2: Stationary Distributions. 

 

4

Tues

Discrete time branching process: HPS 1.6.1:  Computing first passage times.   

 

 

Thurs

Review, Example:  Probability of population extinction.

 

5

Tues

Midterm   

 

 

Thurs

Continuous Time Markov Chains: HPS 3.

 

6

Tues

Continuous Time Markov Chains: HPS 3.

 

 

Thurs

Continuous Time Markov Chains: HPS 3.

 

7

Tues

Continuous Time Markov Chains: HPS 3.

 

 

Thurs

Continuous Time Markov Chains: HPS 3.

 

8

Tues

Continuous Time Markov Chains: HPS 3.

 

 

Thurs

Continuous time Wright Fisher process with mutation. 

 

9

Tues

Hidden Markov Models,

 

 

Thurs

Application of Hidden Markov Models,

 

10

Tues

Review

 

 

 

Grading Policy:

Homeworks (weekly):  40%   Problem sets are assigned every Thursday, due at the beginning of lecture the following Thursday (unless otherwise noted).  Solutions are posted on Friday.  You can be late by at most one day on at most 1 problem set, with deduction of 10% on grade.

 

Midterm (in class): 20%

Final (in class): 40% 

 

Problem Sets:  (Passed out Thursday, due next Thursday, unless otherwise noted.)

 

 

Problem set 1

Problem set 2