10-810 -- Computational Molecular Biology: a machine learning approach

This course focuses on modern machine learning methodologies for computational problems in molecular biology and genetics, including probabilistic modeling, inference and learning algorithms, pattern recognition, data fusion, time series analysis, etc. We will discuss the following biological problems: 1) Analysis of high throughput biological data, such as gene expression data, focusing on issues ranging from data acquisition to pattern recognition and classification. 2) Computational genomics, focusing on gene finding, motifs detection and sequence evolution. 3) Statistical genetics, focusing on the statistical properties of the relationships between traits (phenotypes) and genetic polymorphisms (genotypes), and the analysis of pedigree, SNPS and CGH data. 4) Systems biology, concerning how to combine sequence, expression and other biological data sources (protein-protein interaction, protein-DNA binding and more) to infer the structure and function of different systems in the cell.

Instructors

Ziv Bar-Joseph, WeH 4107, zivbj at cs.cmu.edu

Eric Xing, WeH 4127, epxing at cs.cmu.edu

Course Information

Course structre
Lectures: Monday & Wednesday 15:00-16:20 WEH 4615A
Office Hours, Bar-Joseph: Monday 14:00-15:00, WeH 4107
Office Hours, Xing: Wednesday 16:20-17:20, WeH 4127
Syllabus
Course Material: See assigned reading
Class list: The class list is available for forming project groups.

Problem Sets

Problem set 1 Due on Wednesday, 2/9, in class.

Data set

Problem set 1 solutions

Problem set 2 Due on Wednesday, 3/02, in class.

Data set for problem 3

Data set for problem 4

Problem set 2 solutions

Problem set 3 Due on Monday, 3/28, in class.

Data file from Eisen et al

text files to use for p-value computation: catSize.txt, cin.txt, catNames.txt, geneNames.txt

Matlab file for p-value computation

Problem set 4 Due on Wednesday, 4/06, in class.

Lectures

01/10 Lecture 1: Introduction to molecular biology

01/12 Lecture 2: Statistical modeling of biopolymer sequences

01/19 Lecture 3: The Hidden Markov Models for sequence parsing

01/24 Lecture 4: HMM variants

01/26 Lecture 5: Molecular Evolution and Comperative Genomics

01/31 Lecture 6: Motif Detection

02/09 Lecture 7: Meiosis and recombination

02/14 Lecture 8: 2-point linkage analysis

02/16 Lecture 9: Quantitative trait locus (QTL) mapping

02/21 Lecture 10: SNPs and Haplotype Inference

02/23 Lecture 11: Array CGH

02/28 Lecture 12: Microarrays

03/02 Lecture 13: Normalization

03/14 Lecture 14: Differentially Expressed Genes

03/16 Lecture 15: Clustering expression data

03/21 Lecture 16: Bi-Clustering and Optimal leaf ordering

03/23 Lecture 17: Classification

03/23 Lecture 17a: Classification (cont.)

03/28 Lecture 18: Time series analysis

04/04 Lecture 20: Systems biology

04/06 Lecture 21: Bayesian Networks

04/11 Lecture 22: Graphical models

04/13 Lecture 23: Probabilistic inference in graphical models

04/18 Lecture 24: Physical networks and network motifs

04/20 Lecture 25: Network motifs and protein interactions