Stat 254: Statistics for Computational Biology
Questions about Enrollment?
Please contact
Dean M. Dacumos, MBA
8142A Math Sciences Building
e-mail: dacumos@stat.ucla.edu
Course Outline
The goal of this course is to familiarize the students with
statistical models and approaches that find applications in
computational biology. We will cover some notions of Probability (in order to calculate the significance levels at the base of BLAST), Bayesian inference, Markov Chain Monte Carlo sampling, Expectation Maximization (EM) algorithm, linear regression, model selection, information theory and minimum description length criteria, the bootstrap, False Discovery Rate (FDR) for control of errors in case of multiple comparison.
Each topic will be introduced with specific reference to one problem of computational biology. Articles from the litterature and introductory material will be collected in a course reader.
General information
-
The required textbook is a course reader that is available at the ASUCLA bookstore.
- There will be homework, that will include
writing projects, exercise solving and data analysis. Homework will be typically given every two weeks.
- In order to conduct the data analysis that you will be
required to do, you will need to use R.
This is a free software that you can download from the following site:
The Comprehensive R Archive Network
- The instructor will hold office hours on thursday from
2:30 to 3:30 in Gonda 6357a.
- E-mail is the best way to get in contact with the
instructor.
Her e-mail is
csabatti@mednet.ucla.edu
- Occasionally there will be hand-outs given in class. If you miss
class, it is your responsibility to get a copy of them from your fellow
students. However, extra copies will be outside the instructor office
(Gonda 6357a) and posted on the class web-page when ever possible.
- There is a web-page for the class where homework and
homework solutions will be posted. The address is:
http://www.genetics.ucla.edu/labs/sabatti/Stat254/
Grading plan
The final grade will be a combination of homework scores, final,
and projects.
Prerequisites
This course is designed for students that have a solid foundation in statistics and probability. Stat 100a,b or equivalent are reccommended. Knowledge of bioinformatics problems and approaches as presented in Chem 160/260 is also assumed.