📚 Syllabus & Course Curriculum
Bioinformatics & Computational Biology
Module-by-module breakdown of Explainable AI (XAI) for Single-Cell Multi-Omics Integration, from foundations to a certified capstone project.
Data
• scRNA-seq, scATAC-seq and CITE-seq: what each measures and its sparsity
• Doublets, ambient RNA and dropout as technical confounders
• Paired versus unpaired multi-omic designs
Integration
• Batch correction and integration methods, and their over-correction risk
• Autoencoder and factor-model approaches to joint embedding
• Evaluating integration without a ground-truth alignment
Explainability
• Attribution to genes, peaks and proteins driving a latent dimension
• Attention and gradient methods and their instability on sparse counts
• Distinguishing biological signal from technical covariate
Validation
• Marker-based sanity checks against known cell biology
• Perturbation and held-out validation of proposed mechanisms
• Reporting uncertainty in cell-type and state assignment
Application
• Regulatory inference linking accessibility to expression
• Trajectory and state-transition interpretation, and its assumptions
• Communicating an explainable result to experimental collaborators
Earn government-registered certification in Explainable AI (XAI) for Single-Cell Multi-Omics Integration
e-Certificate and e-Marksheet issued on successful completion.
