BIN504 PROBABILISTIC AND STATISTICAL MODELING FOR BIOINFORMATICS

Course Objectives

The objectives of the course are:

• To expose students to the fundamentals of statistical and probabilistic techniques used in bioinformatics.
  • Statistical issues of biological data
  • Particulars of common bioinformatics analyses
  • Peculiarities of high-throughput experimental data 

• To enable them to apply these techniques on a computational platform 

Course Content

This course will introduce statistical modeling and inference techniques applied to biological problems. The course will cover standard statistical methods, such as multiple regression and principle component analysis, and more recent statistical techniques, such as maximum likelihood methods. Among the techniques covered will be Monte-Carlo-Markov chains using the Metropolis-Hastings algorithm and Gibbs sampling. In addition, the course will cover the use of statistical techniques such as Hidden Markov Models to model family of sequence and structures. Kernel methods and Support Vector Machines for computational biology will also be covered.

Course Learning Outcomes

By the end of the course, students should be able to:

• Model biological experimental data using probabilistic distirbutions
• Have a working knowledge of the R and Python environments
• Be able to design statistically sound high-throughput experiments
• Cluster, visualize, and summarize large amounts of, and highly multidimensional, biodata
• Devise accurate regression models for variables of interest, given this type of biodata