Single-Cell RNA-seq Analysis of NSCLC

From Raw Data to Biological Insights (GSE131907)

This repository provides a complete, beginner-friendly workflow for performing single-cell RNA sequencing (scRNA-seq) analysis using publicly available non-small cell lung cancer (NSCLC) data.

The tutorial demonstrates how raw UMI count matrices can be transformed into meaningful biological insights using reproducible methods implemented in R.

⸻

🎯 Project Goals

The aim of this project is to:

• reconstruct a cellular atlas of NSCLC • identify tumor vs normal transcriptional differences • explore immune and epithelial heterogeneity • perform differential expression analysis • interpret functional pathways associated with tumor progression

⸻

📂 Dataset

GSE131907 – NSCLC Lung Cancer Atlas

Includes: • Tumor lung tissue • Normal lung tissue • Lymph node samples • Effusion samples

Total cells analyzed: ~180,000+ Genes profiled: ~29,000

⸻

🛠 Tools Used

• R • Seurat • ggplot2 • gprofiler2 • patchwork

The workflow avoids heavy dependencies and focuses on stable, reproducible methods.

⸻

🔬 Analysis Workflow

1️⃣ Data Preprocessing

✔ Load UMI matrix ✔ Match cell barcodes with annotation ✔ Create Seurat object ✔ Compute QC metrics: • nFeature_RNA • nCount_RNA • percent mitochondrial reads

⸻

2️⃣ Normalization & Dimensionality Reduction

✔ SCTransform normalization ✔ Identify highly variable genes ✔ PCA for feature reduction ✔ UMAP for visualization ✔ Graph-based clustering

⸻

3️⃣ Cell Type Annotation

Using provided metadata: • Epithelial cells • Myeloid cells • T/NK cells • Fibroblasts • Endothelial cells

⸻

4️⃣ Sub-Atlas Construction

Focused analyses were conducted on:

🧫 Epithelial Cells • Tumor vs normal comparison • Identification of heterogeneous epithelial states • Transcriptional programs associated with tumor progression

🧬 Myeloid Cells • Monocyte/macrophage heterogeneity • Tumor-associated immune remodeling

⸻

5️⃣ Differential Expression Analysis

Comparisons performed: • Tumor vs Normal (Epithelial) • Tumor vs Normal (Myeloid)

Visualizations generated: ✔ Volcano plots ✔ Violin plots ✔ DEG tables

⸻

6️⃣ Functional Enrichment Analysis

Pathway enrichment revealed:

• Immune activation pathways • Inflammatory signaling • Epithelial plasticity • Tumor microenvironment remodeling

⸻

📊 Figures Generated

• UMAP atlas by tissue origin • Cell composition plots • Epithelial sub-atlas • Myeloid clustering • Volcano plots • Functional enrichment plots

All figures are produced in publication-ready format.

⸻

🧠 Key Biological Insights

The analysis demonstrates:

• Tumor epithelial cells exhibit distinct transcriptional programs • Myeloid populations show tumor-associated activation signatures • Immune and inflammatory pathways are enriched in tumor states • Cellular heterogeneity reflects tumor microenvironment dynamics

⸻

🚀 Who Is This Tutorial For?

✔ Beginners in single-cell analysis ✔ Researchers transitioning from bulk RNA-seq ✔ Students learning reproducible genomics workflows ✔ Anyone interested in tumor microenvironment analysis