AI-Guided Single-Cell GRN Analysis & Perturbation

To equip participants with practical skills to use AI-driven GRN inference and single-cell analysis for understanding disease mechanisms and identifying potential regulatory targets.

• Process and analyze scRNA-seq data.
• Infer TF–gene regulatory interactions.
• Identify regulons and master regulators.
• Perform motif-based network refinement.
• Simulate TF knockout or overexpression.
• Analyze and visualize disease-associated GRNs.

🗓️ Day 1: Transcriptomic Data to Gene Regulatory Networks

• Objective: Transform single-cell transcriptomic data into a structured gene regulatory network and infer transcription factor–target relationships using statistical and AI-based methods.
• GRN Fundamentals: Move from differential gene expression analysis toward network-level regulatory inference.
• Data Preparation & QC: Process and quality-control single-cell RNA-seq datasets using AnnData and Scanpy.
• Gene Association Analysis: Explore Pearson, Spearman, and information-based approaches for detecting gene relationships.
• AI-Based GRN Inference: Infer transcription factor–gene relationships using GENIE3 and GRNBoost2.
• Network Construction & Filtering: Build weighted regulatory networks and reduce weak or potentially spurious interactions.

🛠️ Hands-on Lab (Google Colab)

• Task: Preprocess a disease-associated single-cell RNA-seq dataset and generate an initial transcription factor–gene regulatory network.
• Tools Covered: Python, Scanpy, AnnData, GENIE3, GRNBoost2, pandas, and Google Colab.

🗓️ Day 2: TF Motifs, Regulons & Master Regulator Discovery

• Objective: Refine inferred regulatory networks using motif evidence, quantify regulon activity, and identify transcription factors that drive disease-associated cellular programs.
• Regulon Biology: Understand transcription factor–target relationships and the organization of regulatory modules.
• Motif-Enrichment Analysis: Validate candidate TF targets using cis-regulatory motif databases and enrichment analysis.
• Network Pruning: Remove indirect, weak, or unsupported regulatory interactions to improve GRN specificity.
• Regulon Activity: Calculate cell-level transcription factor and regulon activity using AUCell.
• Master Regulators: Identify disease-associated transcription factors and dominant regulatory programs across cell populations.

🛠️ Hands-on Lab (Google Colab)

• Task: Run a pySCENIC workflow, identify high-confidence regulons, calculate regulon activity, and generate transcription factor activity heatmaps.
• Tools Covered: pySCENIC, AUCell, cisTarget databases, Scanpy, Python, and Google Colab.

🗓️ Day 3: In Silico Perturbation & GRN Visualization

• Objective: Interpret regulatory networks, simulate transcription factor perturbations, and visualize disease-relevant regulatory changes for mechanistic and therapeutic research.
• Network Interpretation: Identify regulatory hubs, bottlenecks, master regulators, and candidate therapeutic targets.
• Gene Perturbation: Simulate transcription factor knockout and overexpression using CellOracle.
• Cell-State Prediction: Estimate how regulatory perturbations may redirect cellular states and trajectories.
• Network Analysis: Quantify connectivity, centrality, regulatory rewiring, and network-level changes using NetworkX.
• Publication Visualization: Export and visualize gene regulatory networks using Cytoscape-compatible GraphML and JSON formats.

🛠️ Hands-on Lab (Google Colab)

• Task: Simulate TF knockout or overexpression, compare baseline and perturbed regulatory networks, identify key regulatory shifts, and export publication-ready GRN visualizations.
• Tools Covered: CellOracle, NetworkX, Cytoscape, Scanpy, GraphML, JSON, Python, and Google Colab.

• Ph.D. scholars and researchers
• Bioinformaticians and computational biologists
• Molecular and cell biologists
• Genomics and transcriptomics researchers
• Systems biology researchers
• Cancer, neuroscience, immunology, and disease researchers
• Drug discovery and biotechnology professionals
• Faculty and academicians working with single-cell or omics data

• Build GRNs from single-cell transcriptomic data.
• Identify key transcription factors and regulatory programs.
• Compare regulatory activity across cell states.
• Predict effects of gene perturbations.
• Discover potential disease mechanisms and therapeutic targets.
• Generate publication-ready GRN visualizations.