Search Modules > CS4220 > Module Overview

Learning Outcomes

TopPresent-day biomedical researchers are confronted by vast amounts of data from genome sequencing, microscopy, high-throughput analytical techniques for DNA, RNA, and proteins, and a host of other new experimental technologies. Coupled with the advances in computing power, this flow of information should enable scientists to model and understand biological systems in novel ways.

The goals of CS4220 (Knowledge Discovery Methods in Bioinformatics) are:

Expose students to knowledge discovery techniques,
Enhance students' flexible and logical problem solving skills,
Develop students' understanding, in depth, of bioinformatics and issues in analysis of real-life high-throughput biological data.

To achieve these goals, we do a series of in-depth studies and hands-on projects on topics such as gene expression profile analysis, proteomic profile analysis, epistatic interaction detection, protein family recognition, etc.

At the end of the course, students will be able to identify the relevant knowledge discovery techniques for different biological data to uncover new information, as well as be confident in formulating and validating hypothesis underlying observations from biological data.

Prerequisites

TopThe student is assumed to have already taken CS2220, and has an appreciation of computer algorithms, computer programming, and basic molecular biology.

Teaching Modes

TopThe teaching will primarily be by lectures. Each lecture will focus on one topic, and will typically begin with a simple introduction to the topic, a review of current approachs on the topic, and at least one in-depth case study on the topic. To increase students' appreciation on a topic, a list of must-read and good-to-read articles will be provided for students to read before each class. Students are especially expected to have read the must-read articles before hand. For some topics, students may also be asked to present and discuss the articles read.

Syllabus

TopThe course comprises the following units:

Essence of Biostatistics

Basics of biostatistics
Statistical estimation
Hypothesis testing
Measurement data: z-test, t-test
Categorical data: c2-test, fisher’s exact test
Non-parametric methods

Essence of Data Mining

Clustering
Association rules
Classification
Class-imbalance learning

Gene Expression Profile Analysis

Basic gene expression profile analysis
Common issues
Batch effect & normalization
Improving reproduciblity, sensitivity, and precision

Proteomic Profile Analysis

Basic proteomic profile analysis
Common issues
Improving consistency
Improving coverage

Protein Interaction Network

Overview of biological networks
Use of biological networks in enhancing bioinformatics analysis
Consistency, comprehensiveness, and compatibility of biological pathway databases
Integration of pathway databases
Reliability of PPIN
Identifying noise and missing edges in PPIN
An advanced example on quality assessment of PPIN

Protein Complex Prediction

Overview of protein complex prediction
A case study: MCL-CAw
Impact of PPIN cleansing
Detecting overlapping complexes
Detecting low-density complexes
Detecting small complexes

Protein Function Prediction

Basic protein function prediction
"Guilt by association" of other properties
Protein function prediction from PPIN
"Guilt by association" of multiple types of information

Pathway perturbations in a disease context

Assessment

TopThere will be 3-4 assignments, 1 project, and a final exam. The assignments and project will collectively contribute upto 60% of the course grade; the final exam will contribute upto 40% of the course grade.

Workload

Top2-1-0-4-3

Workload Components : A-B-C-D-E
A: no. of lecture hours per week
B: no. of tutorial hours per week
C: no. of lab hours per week
D: no. of hours for projects, assignments, fieldwork etc per week
E: no. of hours for preparatory work by a student per week