Computational Neurobiology with AI: Protein Dynamics, Biomarkers, Molecular Pathways & Drug Discovery

This three-day workshop introduces participants to computational and AI-driven approaches used to study neurological and neurodegenerative diseases. The program begins with protein structure, mutation effects, and disease-associated molecular changes, then progresses to transcriptomics, biomarker discovery, pathway enrichment, and network biology. Participants will finally explore therapeutic target identification, drug repurposing, molecular docking, ADMET assessment, and AI-assisted candidate prioritization through guided hands-on workflows.

To provide participants with an integrated understanding of computational neurobiology by connecting protein-level disease mechanisms, biomarker discovery, pathway and network analysis, and AI-assisted therapeutic discovery for neurological and neurodegenerative disorders.

• Understand molecular and protein-level mechanisms associated with neurological disorders.
• Analyze disease-associated proteins, mutations, structural changes, and stability.
• Explore neurobiological datasets for biomarker discovery.
• Perform differential expression and disease-signature analysis.
• Apply pathway enrichment and network biology approaches.
• Identify hub genes, molecular drivers, and therapeutic targets.
• Use AI/ML approaches for biomarker and disease-pattern analysis.
• Explore drug repurposing and computational compound screening.
• Evaluate drug candidates using docking, drug-likeness, and ADMET parameters.
• Integrate structural, omics, network, and drug-discovery evidence into a unified workflow.

Day 1: Protein Dysfunction, Structural Dynamics & Disease Mechanisms

• Protein folding, misfolding, aggregation, and intrinsically disordered proteins in neurodegenerative disease. Disease-associated proteins in Alzheimer’s disease, Parkinson’s disease, ALS, and related disorders . Sequence, domain, and structural analysis using experimental and predicted protein information. Identification of disordered and aggregation-prone regions . Disease-associated variant interpretation and mutation-impact analysis. AlphaFold-based structure interpretation, confidence assessment, and limitations for intrinsically disordered proteins. Concepts of protein stability, conformational dynamics, oligomerization, and molecular dynamics.
• Tools: UniProt, RCSB PDB, AlphaFold DB, ClinVar, AlphaMissense, IUPred2A/AGGRESCAN, PyMOL/ChimeraX
• Hands-on: Analyze a disease-associated protein for structural confidence, disorder and aggregation propensity, disease variants, and mutation-related effects.

Day 2: Candidate Biomarkers, Molecular Pathways & Network Biology

• Explore public neurobiology and neurodegenerative-disease transcriptomic datasets. Perform differential gene-expression analysis and identify disease-associated molecular signatures. Visualize disease patterns using PCA and UMAP . Perform functional enrichment and biological pathway analysis. Build protein–protein interaction networks and identify hub genes and molecular drivers. Prioritize candidate biomarkers using statistical and biological evidence. Apply machine-learning approaches such as Random Forest/XGBoost for biomarker ranking. Interpret important predictors using SHAP concepts . Overview of single-cell transcriptomics and its applications in neurological disease research.
• Tools: GEO, GEO2R/DESeq2, Python/R, pandas, scikit-learn, UMAP, Enrichr, STRING, Cytoscape, XGBoost, SHAP
• Hands-on: Identify differentially expressed genes, enriched pathways, network hubs, and ML-ranked candidate biomarkers from a neurobiology dataset.

Day 3: Therapeutic Target Prioritization & AI-Assisted Drug Discovery

• Translate protein, transcriptomic, pathway, and network evidence into therapeutic target candidates . Prioritize structurally suitable and biologically relevant drug targets. Retrieve and evaluate known compounds for drug repurposing and small-scale computational screening . Analyze drug–target interactions and molecular descriptors. Assess drug-likeness and pharmacokinetic properties. Perform ADMET prediction for candidate compounds. Conduct molecular docking using structurally characterized targets. Analyze binding interactions and docking poses. Integrate docking, ADMET, molecular descriptors, and biological evidence for candidate ranking. Apply AI-assisted approaches for therapeutic-candidate prioritization.
• Tools: RCSB PDB, PubChem, ChEMBL/BindingDB, AutoDock Vina or CB-Dock2, PyMOL, SwissADME, Python
• Hands-on: Prioritize a therapeutic target, screen selected compounds, evaluate docking and ADMET results, and rank candidate therapeutics using integrated computational evidence.

• Analyze disease-associated proteins and structural alterations.
• Interpret mutation and protein-stability predictions.
• Identify candidate biomarkers from neurobiological datasets.
• Perform pathway enrichment and protein–protein interaction analysis.
• Recognize key disease-associated genes, proteins, and molecular pathways.
• Apply basic machine-learning approaches for biomarker prioritization.
• Identify and prioritize potential therapeutic targets.
• Retrieve and evaluate candidate compounds for neurological disorders.
• Perform basic molecular docking and interaction analysis.
• Assess drug-likeness and ADMET properties.
• Build an integrated workflow fromdisease mechanism to biomarker, therapeutic target, and drug candidate prioritization.

Biotechnology and Life Sciences students Neuroscience and Neurobiology researchers Bioinformatics and Computational Biology learners Molecular Biology and Biochemistry researchers Pharmacology and Drug Discovery researchers Biomedical Science students and professionals Genetics and Genomics researchers Pharmaceutical Sciences students PhD scholars and postdoctoral researchers Faculty members and research professionals