📚 Syllabus & Course Curriculum
Bioinformatics & Computational Biology
Module-by-module breakdown of AI in Genomics & Personalized Medicine, from foundations to a certified capstone project.
Data Foundations
• Platform characteristics and their error profiles: short read, long read and their trade-offs
• FASTQ, BAM and VCF: what each stores and where information is lost
• Alignment and variant calling with GATK best practices
• Quality control gates that must pass before any model sees the data
Interpretation
• Annotation with VEP or ANNOVAR; transcript choice changes the answer
• Population frequency filtering with gnomAD and ancestry-matched controls
• ACMG/AMP classification criteria and evidence weighting
• In-silico predictors including CADD, REVEL and AlphaMissense, and how much to trust them
Machine Learning
• Feature engineering on variant, expression and multi-omic data
• Class imbalance in rare-disease and biomarker problems
• Cross-validation that respects population structure and relatedness
• Data leakage: the most common reason a genomics model fails in the clinic
Deep Learning
• CNNs for regulatory element and chromatin prediction
• DeepVariant and learned variant calling
• Long-range architectures for expression prediction from sequence
• Splice-effect prediction with SpliceAI and interpreting its scores
Clinical Genomics
• Constructing polygenic risk scores and their poor portability across ancestries
• Pharmacogenomic star-allele calling and CPIC guideline implementation
• Embedding results in clinical decision support without alert fatigue
Governance
• Analytical and clinical validation of an AI-assisted genomic test
• Explainability requirements when a model informs a clinical decision
• Ancestry representation and how unrepresentative training data harms patients
• Consent, data governance and compliance under GDPR and India's DPDP Act
Earn government-registered certification in AI in Genomics & Personalized Medicine
e-Certificate and e-Marksheet issued on successful completion.
