BMS5302 · Machine learning: AI for bioinformatics

Overview

In this unit, you will explore the application of machine learning methods in bioinformatics, including both theoretical aspects and practical implementation. You will be introduced to the conceptual foundations and appropriate usage of algorithms for dimensionality reduction, clustering, classification and prediction as applied to the analysis of multi-omics and imaging data from biological studies. You will become familiar with sound practices for all stages of machine learning analyses, including data pre-processing and feature selection through to model development, validation, and interpretation. You will learn to critically evaluate model performance, define the limitations of the algorithms used and interpret model predictions for both technical and non-technical audiences. A special emphasis will be placed throughout on how to select and implement machine learning approaches suitable to the biological context of a given research question. This unit will develop your conceptual understanding and practical experience in applying machine learning techniques to complex biological data through online lectures, hands-on workshop sessions, and independent project work. By the end of the unit, you will be able to knowledgeably apply and evaluate the performance of machine learning algorithms to common analysis problems in bioinformatics.

Offerings

Assessment

The Handbook does not list a final examination among the assessment items. That is not a guarantee there is none.

Assessment details may change. Please refer to the assessment information in Moodle closer to the start of the teaching period.

Requisites

Learning outcomes

1. Differentiate and categorize machine learning methods commonly used in bioinformatics;
2. Formulate biological data analysis problems amenable to machine learning approaches;
3. Collect and prepare suitable input for machine learning algorithms;
4. Design and construct an analysis based on machine learning to address a biological problem;
5. Justify choice of machine learning algorithms to address defined biological questions;
6. Evaluate, interpret and communicate machine learning predictions for non-specialist audiences.

Workload

Average of 6 hours teacher-directed learning / week (on-campus workshops, online learning materials) plus 6 hours student-directed learning. Total per week = 12 hours

Student experience, not official rules. Nothing here changes what the Handbook says.