You are invited to this FREE 2.5 hour workshop.
AI-Driven Insights in Spatial Biology: Integrating Spatial Transcriptomics and Multi-Omics
Workshop Leader

Arvind Rao, Professor of Computational Medicine and Bioinformatics, University of Michigan Ann Arbor
Audience: Professionals in life sciences, pharma, and biotech with limited experience in AI and ML.
Course Objectives:
- Understand the role of AI, ML, and deep learning in spatial biology and spatial transcriptomics.
- Explore integration of multi-omics data from bulk, single-cell, and spatial transcriptomics.
- Gain insights into foundation models and LLM (Large Language Model) agentic workflows for analysis.
- Learn through practical case studies with actionable applications.
Course Agenda:
1. Introduction to Spatial Biology and Transcriptomics (20 min)
• Importance of spatial context in biological discovery.
• Overview of technologies: spatial transcriptomics, CODEX, and
multiplexed imaging.
• Applications in pharma: target discovery, biomarker identification, and patient stratification.
2. AI, ML, and DL in Spatial Transcriptomics and CODEX (40 min)
• Intuitive overview of AI and ML techniques: clustering, classification, and feature extraction.
• Role of deep learning in image-based segmentation and spatial relationships.
• Case study: Using AI to identify tumor-immune interactions.
3. Integration of Multi-Omics Data (30 min)
• Combining bulk, single-cell, and spatial transcriptomics for enriched analysis.
• Techniques for spatial deconvolution and cell-type mapping.
• Knowledge graphs to incorporate biological context.
• Case study: Multi-modal integration for immune cell profiling.
4. Foundation Models and LLM Agentic Workflows (40 min)
• Introduction to foundation models: tissue segmentation and spatial feature extraction.
• LLM workflows for automated spatial transcriptomics analysis.
• Future applications: integrating multi-omics and dynamic querying.
5. Wrap-Up and Future Directions (20 min)
• Emerging trends: multi-modal models and advanced AI techniques.
• Ethical considerations: transparency, reproducibility, and interpretability.
• Q&A and open discussion.