Introduction to Machine Learning in Bioinformatics

This course teaches students how to perform robust machine learning analysis for biomedical molecular data, covering preprocessing, model choice, supervised and unsupervised learning, model evaluation, interpretability, and project-based scientific reporting.

Course abbreviation: ML-BIO

Instructor: Dr. Ahmed Osman

Bioinformatics Master Student at UdS; Research Assistant at Integrative Cellular Biology Department; Former Lead Data Scientist at Vodafone.

Workload: 12 lectures, 3 hours each. Total workload: 42 hours: 36 hours of lectures and tutorials and 6 hours of self studies.

Entrance requirements: Basic knowledge of biology and computer science.

Used media: PowerPoint presentation

Objectives

• Understand the theoretical basics of machine learning
• Apply machine learning to biomedical and omics data
• Preprocess and clean biomedical data for ML analysis
• Build and evaluate predictive models
• Interpret machine learning models and identify important biomarkers

Competences to be Developed

• Python scripting and packages such as PyTorch and scikit-learn
• Data-specific cleaning and preprocessing pipelines
• Unsupervised and supervised learning methods
• Adequate model evaluation based on data size, quality, and distribution
• Scientific interpretation, writing, and discussion of results

Assessment

• Finalize a statistical learning project
• Choose and apply appropriate methods for a specified problem
• Compile outcomes as a scientific research article
• Present, discuss, and review projects in the last lecture