Programming Biology with Foundation Models and Agentic AI
Program Intelligent Biology with Foundation Models and Autonomous AI
Skills you will gain:
About Program:
Foundation models trained on massive biological datasets—such as protein sequences, DNA/RNA, structural databases, and multi-omics repositories—are transforming bioinformatics and molecular research. Models inspired by large language models (LLMs) can now predict protein structure, annotate genomes, design sequences, and extract biological meaning from complex datasets. These models reduce the need for task-specific training and enable transfer learning across biological domains.
Agentic AI extends this capability by building autonomous systems that can reason, plan, query databases, run code, and iteratively refine results. In this workshop, participants will learn how to integrate foundation models with Python-based programming, APIs, and tool-using agents to build intelligent biological analysis pipelines. The focus is entirely dry-lab, emphasizing reproducible computational workflows for genomics, proteomics, structure prediction, and systems biology.
Aim:
This workshop aims to train participants in using biological foundation models and agentic AI systems to solve complex problems in modern biology. It focuses on programming workflows that integrate large pre-trained models for sequences, structures, and multi-omics data. Participants will learn to design AI-driven pipelines that automate hypothesis generation, analysis, and interpretation. The program bridges computational biology, generative AI, and autonomous research systems.
Program Objectives:
- Understand biological foundation models (sequence, structure, multi-omics).
- Learn to program workflows integrating pre-trained biological AI models.
- Build agentic AI systems that query databases and execute bioinformatics tasks.
- Apply transfer learning and prompt-based strategies for biological problems.
- Evaluate model outputs with biological and computational validation methods.
What you will learn?
Day 1: Foundation Models for Genetic & Molecular Representation Learning
- DNA, RNA, and protein as language, Tokenization strategies in genomics
- Embeddings vs handcrafted biological features, Transfer learning in biological systems
- ESM, ProtT5, DNABERT, Multi-omics transformers
- In-context learning in biological systems
- Limitations of pretrained biological models
- SMILES encoding, Graph Neural Networks for molecules
- Cross-modal embeddings (sequence ↔ structure ↔ function)
- Functional similarity via embedding distance
Hands-on Tools & Platforms
- HuggingFace Transformers
- ESM-2 protein model
- PyTorch
- Google Colab
- Matplotlib / Seaborn
Day 2: Generative AI & Agentic Systems for Drug Discovery
- Molecule optimization workflows
- Sequence-to-function modeling
- CRISPR guide optimization
- Scaffold modification strategies
- AI agents for literature search
- Docking tool calls (conceptual demonstration)
- Automated data retrieval
- Reinforcement learning from human feedback (RLHF concept)
- Designing perturbation experiments
- Hypothesis ranking using LLMs
- Automating biological reasoning workflows
Hands-on Tools & Platforms
- RDKit
- PyTorch
- LangChain
- OpenAI / Open-source LLM APIs
- PubMed API (demonstration)
- Matplotlib
Day 3: Multiscale Modeling, Causality & AI Virtual Cells
- Multi-omics integration
- Spatial transcriptomics modeling
- Systems-level embeddings
- Causal graphs in biology
- Counterfactual reasoning
- Mechanism vs correlation
- Interpretable AI for biomedical systems
- Simulating disease progression
- Drug response modeling
- Dynamic biological systems
- Toward AI-powered digital twins
Hands-on Tools & Platforms
- NetworkX
- PyTorch
- Google Colab
- MC Dropout
- Ensemble prediction methods
- Matplotlib / Seaborn
Mentor Profile
Fee Plan
Get an e-Certificate of Participation!
Intended For :
- Doctoral Scholars & Researchers: PhD candidates seeking to integrate computational workflows into their molecular research.
- Postdoctoral Fellows: Early-career scientists aiming to enhance their data-driven publication profile.
- University Faculty: Professors and HODs interested in modern bioinformatics pedagogy and tool mastery.
- Industry Scientists: R&D professionals from the Biotechnology and Pharmaceutical sectors transitioning to genomic-driven discovery.
- Postgraduate Students: Final-year PG students looking for specialized research-grade exposure beyond standard curricula.
Career Supporting Skills
Program Outcomes
Participants will be able to:
- Use foundation models for sequence and structure analysis.
- Build AI-powered bioinformatics workflows using Python.
- Design agent-based systems for automated biological research tasks.
- Integrate multiple biological data types using AI reasoning.
- Develop reproducible pipelines for research and industry applications.