Workshop Structure

Course Content

1. Day 1 - Theoretical Foundations

   • Morning Session:

   • Introduction to Single-Cell Technologies

     • - Evolution and need for single-cell analysis

     • - Comparison of platforms (10x Genomics, SMART-Seq, Drop-Seq)

     • - Applications: cancer, immunology, developmental biology

   • Experimental Design & Protocols

     • - Sample collection, dissociation, viability check

     • - Barcoding, reverse transcription, library preparation

     • - Avoiding batch effects: experimental best practices

   • scRNA-seq Data Workflow Overview

     • - From FASTQ to expression matrix

     • - Introduction to Cell Ranger

   • Break

   • Afternoon Session:

   • Quality Control

     • Ambient RNA: CellBender

     • Doublets: Scrublet

     • Mitochondrial % and QC thresholds

     • Denoising: brief on MAGIC and imputation caveats

   • Normalization & Feature Selection

     • Log-normalization as the standard workflow

     • HVG selection: mean-variance & VST

     • Biological relevance of features

   • Dimensionality Reduction & Clustering

     • PCA → UMAP/t-SNE

     • Graph-based clustering (KNN, Louvain)

     • Interpretation and pitfalls

   • Cell Type Annotation

     • Marker-based + reference mapping

     • Tools: SingleR, Azimuth

     • Manual curation tips

2. Day 2 – Hands-On Practical

   • Morning Session:

   • Cell Ranger Output Interpretation (R & command line)

     • - Load filtered feature-barcode matrix

     • - Inspect Cell Ranger output structure and QC metrics (JSONs, HTML)

   • Preprocessing & QC (R: Seurat, Python: Scanpy demo)

     • - Filter cells & genes by count, gene, and mitochondrial thresholds

     • - Apply ambient RNA correction with CellBender (Python demo)

     • - Detect doublets with Scrublet (Python) or DoubletFinder (R)

     • - Generate visual QC plots

   • Normalization & HVG Selection (R main)

     • - Apply Log-normalization

     • - Identify and visualize HVGs

     • - Assess biological vs technical variation

   • Lunch Break

   • Afternoon Session

   • Dimensionality Reduction & Clustering (R main)

     • - Perform PCA

     • - Construct KNN graph and apply clustering (Louvain)

     • - Visualize with UMAP

   • Cell Type Annotation (R main, Python optional)

     • - Visualize marker gene expression

     • - Automatic annotation with Azimuth or SingleR

     • - Manual curation and cluster naming

   • Dataset Integration

     • - Apply Harmony (R) to correct for batch effects

     • - Compare with scVI (Python) for latent variable modeling

     • - Discussion on strengths and use cases for each method

   • Differential Expression Analysis

     • - Compare resistant vs sensitive mice to PARPi therapy

     • - Identify DE genes across key immune and tumor populations

     • - Discuss biological implications in the context of the publication

Who Should Enroll?

• If you are a first-year undergraduate student, this workshop is NOT for you. You CANNOT attend.

• - Researchers familiar with bulk RNA-seq who want to transition to single-cell analysis.

• - Biologists and bioinformaticians interested in exploring scRNA-seq technologies and workflows.

• - Undergraduate students (second year or above) with a strong interest in genomics and data analysis.

• - Master’s and PhD students working in genomics, immunology, cancer biology, or related fields

• - Computational scientists aiming to apply R/Python skills in a cutting-edge biomedical context

• - Anyone curious about real-world applications of scRNA-seq in cancer and immunotherapy research

• No prior single-cell experience is required, but basic understanding of transcriptomics is highly recommended.