📚 Syllabus & Course Curriculum
Bioinformatics & Computational Biology
Module-by-module breakdown of Machine Learning for Cheminformatics and Genomics, from foundations to a certified capstone project.
Chemical Data
• Fingerprints, descriptors and learned graph representations
• Data curation: salts, tautomers, duplicates and activity cliffs
• Scaffold and time splits instead of random splits
Genomic Data
• Encoding sequence, variants and expression for models
• Population structure and relatedness as leakage sources
• Dimensionality and regularisation in omics-scale feature spaces
Models
• Tree ensembles as strong baselines on tabular chemical and genomic features
• Graph neural networks for molecules and interaction networks
• Multi-task and transfer learning across related endpoints
Reliability
• Applicability domain estimation for chemical models
• Conformal prediction and calibrated uncertainty
• Recognising when a prediction should not be acted on
Integration
• Target-ligand modelling and proteochemometrics
• Drug response prediction from cell line omics
• Prospective validation and the gap from benchmark to bench
Earn government-registered certification in Machine Learning for Cheminformatics and Genomics
e-Certificate and e-Marksheet issued on successful completion.
