AI-Guided Targeted Protein Degradation: PROTACs, Molecular Glues & Degrader Design
From Protein Inhibition to Protein Destruction — Design, Model and Prioritise Next-Generation Degraders with AI.
About This Course
This three-day hands-on workshop introduces participants to Targeted Protein Degradation (TPD), an emerging therapeutic strategy that removes disease-associated proteins rather than simply inhibiting them. Participants will explore PROTACs, molecular glues, E3-ligase biology, ternary-complex formation, linker design and AI-assisted degrader prediction through guided computational activities using structural biology, cheminformatics and machine-learning tools.
Aim
To provide participants with practical knowledge of targeted protein degradation, PROTAC and molecular-glue design, ternary-complex modelling, molecular-property analysis and AI-assisted degrader prioritisation for modern drug-discovery and therapeutic-research applications.
Workshop Objectives
- Understand the biological principles of targeted protein degradation and event-driven pharmacology.
- Explore the ubiquitin–proteasome system and the role of E3 ligases in protein degradation.
- Understand PROTAC architecture, including target ligands, linkers and E3-ligase recruiters.
- Identify disease-relevant protein targets and suitable ligands for degrader development.
- Analyse target–PROTAC–E3 ternary-complex structures.
- Evaluate linker and molecular properties influencing degrader performance.
- Understand molecular-glue degraders and induced-proximity mechanisms.
- Generate molecular descriptors and fingerprints for degrader molecules.
- Build machine-learning models for degrader-activity prediction.
- Apply explainable AI and multi-parameter ranking for candidate prioritisation.
Workshop Structure
📅 Day 1: Targeted Protein Degradation & PROTAC Design
Focus: Understanding TPD biology and building a computational PROTAC design strategy.
Topics Covered
- Event-driven protein degradation and the ubiquitin–proteasome system
- Role of E3 ligases, including CRBN and VHL
- PROTAC architecture: warhead, linker and E3-ligase recruiter
- Target selection, ligand identification and E3-ligase pairing
- Introduction to molecular glues and other induced-proximity modalities
🛠️ Hands-on Activities
- Select a disease-relevant protein target
- Retrieve target and E3-ligase structures
- Identify suitable ligands and recruiter molecules
- Build a preliminary PROTAC design strategy
🧰 Tools Covered:
UniProt, RCSB PDB, AlphaFold DB, PubChem, ChEMBL, PROTAC databases, RDKit
📅 Day 2: Ternary Complex Modeling, Linker Design & Molecular Glues
Focus: Evaluating structural features that influence productive protein degradation.
Topics Covered
- Target–PROTAC–E3 ternary-complex formation
- Ternary-complex geometry and cooperativity
- Linker length, flexibility, polarity and attachment points
- PROTAC physicochemical and beyond-Rule-of-Five properties
- Molecular glue degraders and neosubstrate recruitment
🛠️ Hands-on Activities
- Analyse a representative ternary complex
- Compare alternative linker designs
- Calculate key molecular properties
- Visualise and prioritise degrader architectures
🧰 Tools Covered:
RCSB PDB, AlphaFold DB, RDKit, PyMOL, Py3Dmol, Python, PROTAC databases
📅 Day 3: AI-Based Degrader Prediction & Candidate Prioritisation
Focus: Using machine learning and explainable AI to predict degrader performance and rank candidates.
Topics Covered
- Key TPD metrics: DC₅₀, Dmax, potency and selectivity
- Degrader descriptors and molecular fingerprints
- Machine-learning models for activity prediction
- Explainable AI for identifying important molecular features
- Multi-parameter ranking using activity, structure and physicochemical properties
- Computational limitations and experimental-validation considerations
- 🛠️ Hands-on Activities
- Prepare a degrader activity dataset
- Generate RDKit descriptors and fingerprints
- Train and evaluate an ML prediction model
- Apply SHAP analysis
- Rank degrader candidates using multiple criteria
🧰 Tools Covered:
RDKit, Python, Pandas, Scikit-learn, XGBoost, SHAP, PyMOL, Py3Dmol
Who Should Enrol?
- Graduate and Postgraduate Students from Biotechnology, Bioinformatics, Biochemistry, Molecular Biology, Pharmacy, Pharmaceutical Sciences, Chemistry, Biomedical Sciences, Life Sciences, and related disciplines
- Ph.D. Scholars and Researchers working in drug discovery, cancer biology, protein science, medicinal chemistry, molecular pharmacology, structural biology, chemical biology, or computational biology
- Academicians and Faculty Members interested in targeted protein degradation, PROTACs, molecular glues, AI-enabled drug discovery, and computational therapeutic research
- Industry Professionals from pharmaceutical, biotechnology, CRO, medicinal chemistry, computational chemistry, bioinformatics, and AI/ML-driven drug-discovery teams
Important Dates
Registration Ends
September 21, 2026
IST 4: 30 PM
Workshop Dates
September 21, 2026 – September 23, 2026
IST 5:30 PM
Workshop Outcomes
- Explain how targeted protein degradation differs from conventional drug inhibition.
- Identify suitable disease targets and E3 ligases for degrader development.
- Design the basic architecture of a PROTAC molecule.
- Interpret target–PROTAC–E3 ternary complexes.
- Evaluate linker and molecular properties affecting degrader behaviour.
- Understand the mechanism of molecular-glue degraders.
- Generate molecular descriptors and fingerprints for degrader molecules.
- Build and evaluate an ML model for degrader-activity prediction.
- Use explainable AI to identify important molecular features.
- Prioritise degrader candidates using structural, molecular and predicted activity information.
- Prepare a computationally supported degrader-design strategy for further research or experimental validation.
Fee Structure
Student Fee
₹2299 | $65
Ph.D. Scholar / Researcher Fee
₹3499 | $80
Academician / Faculty Fee
₹4999 | $100
Industry Professional Fee
₹5499 | $120
What You’ll Gain
- Live & recorded sessions
- e-Certificate upon completion
- Post-workshop query support
- Hands-on learning experience
View All Feedbacks →