Harness AI/ML to solve genomic challenges using R.
Aim :
This workshop equips researchers with practical skills to apply AI/ML to genomic data using R. Covering foundational to advanced topics (PRS, interpretable AI, generative models), participants will learn to process NGS data, build predictive models, and address ethical considerations. Through hands-on sessions with cutting-edge R packages (tidymodels, MOFA2), attendees will gain experience in transforming raw genomic data into biological insights. The workshop emphasizes real-world applications, from variant interpretation to multi-omics integration, preparing researchers for the AI-driven future of genomics.
Job Opportunity :
Genomic Data Scientist (biotech/pharma)
Clinical Genomics ML Specialist
Bioinformatics Engineer (AI focus)
Genomic AI Product Developer
Research Scientist (computational genomics)
Objective :
Preprocess NGS data for ML using R/Bioconductor
Implement polygenic risk score models with modern ML
Apply explainable AI techniques to genomic predictions
Detect and mitigate bias in genomic ML models
Integrate multi-omics datasets using dimensionality reduction
Interface R with deep learning frameworks for genomics
Deploy genomic ML models via Shiny applications
Duration :3 Days (1.5 Hours/Day) @ IST 08:00 PM daily
Start Date :14/06/2025 -
End Date :16/06/2025
Product Structure :
Day 1: Foundations of AI/ML in Genomics
Introduction to AI in Genomics
Overview of latest trends: Transformers in genomics (DNABERT, Nucleotide Transformer)
R vs. Python: When to use R for genomics ML (BioConductor, tidyomics)
Data Preprocessing & Feature Engineering
Handling NGS data in R: plyranges, VariantAnnotation
Creating ML-ready datasets from VCFs/FASTQs
Hands-on:
Building a simple variant predictor using tidymodels
Day 2: Advanced ML for Genomics
Polygenic Risk Scores (PRS) with ML
Latest methods: PRS-CSx, transformer-based PRS
Implementing in R with glmnet + xgboost
Interpretable AI for Genomics
SHAP values for variant interpretation (DALEX in R)
Detecting bias in genomic ML models
Hands-on:
Training an explainable PRS model
Day 3: Cutting-Edge Applications
Generative AI in Genomics
AI for synthetic DNA design (using R interfaces to PyTorch)
Ethical considerations
Multi-Omics Integration
Combining RNA-seq, methylation, proteomics with MOFA2
Live Demo:
Deploying a genomic ML model as a Shiny app
Panel Discussion:
Future of AI in genomics (industry + academia perspectives)
Benefits to the participants :
Master in-demand skills at AI-genomics intersection
Learn to implement 2024’s leading genomic ML methods
