A comprehensive single-cell RNA sequencing analysis pipeline for breast cancer using Seurat, SingleR, Gene Ontology, KEGG, Reactome, and Tumor Microenvironment characterization.
This project performs a comprehensive single-cell RNA sequencing (scRNA-seq) analysis to characterize cellular heterogeneity and the tumor microenvironment in breast cancer.
The workflow integrates quality control, dimensionality reduction, unsupervised clustering, cell type annotation, differential expression analysis, and functional pathway analysis to identify biologically meaningful cell populations and molecular signatures.
GEO Accession: GSE176078
Organism: Homo sapiens
Disease: Breast Cancer
Technology: Single-cell RNA Sequencing
Dataset contains:
- Gene expression count matrix
- Cell barcode information
- Gene annotation
- Cell metadata
Analysis was performed using R, Seurat, SingleR, and Bioconductor packages.
Raw scRNA-seq Data
↓
Quality Control & Filtering
↓
Normalization
↓
Highly Variable Gene Identification
↓
PCA Dimensionality Reduction
↓
UMAP Visualization
↓
Cell Clustering
↓
Marker Gene Identification
↓
Cell Type Annotation (SingleR)
↓
Differential Expression Analysis
↓
GO Biologocal Process Enrichment
↓
Reactome Pathway Analysis
↓
Tumor Microenvironment Characterization
↓
Immune Landscape Analysis
Cells were filtered based on:
nFeature_RNA > 300
nFeature_RNA < 7000
Mitochondrial percentage < 15%
Quality metrics evaluated:
- Number of detected genes per cell
- Total RNA counts
- Mitochondrial gene expression
Highly variable genes were identified followed by PCA-based dimensionality reduction.
The optimal number of principal components was selected using an elbow plot.
Unsupervised clustering identified distinct cellular populations within the breast cancer microenvironment.
Cell identities were assigned using:
SingleR reference-based annotation
Identified cell populations included:
| Cell Population | Description |
|---|---|
| T Cells | Adaptive immune population |
| NK Cells | Cytotoxic lymphocytes |
| B Cells | Antibody-producing immune cells |
| Macrophages | Tumor-associated immune cells |
| Monocytes | Myeloid immune population |
| Endothelial Cells | Tumor vasculature |
| Fibroblasts | Stromal population |
| Epithelial Cells | Tumor-associated cells |
Cluster-specific marker genes were identified using differential expression analysis.
Example markers:
| Cell Type | Representative Markers |
|---|---|
| T Cells | CD3D, CD3E, CD2, CCL5 |
| NK Cells | NKG7, GNLY |
| B Cells | CD79A, MS4A1 |
| Macrophages | LST1, TYROBP |
| Endothelial Cells | VWF, EMCN, PLVAP |
| Fibroblasts | COL1A1, COL3A1 |
Cluster-specific differential expression analysis was performed to identify genes enriched in each cellular population.
Criteria:
Adjusted P-value < 0.05
Positive log2 Fold Change
Results:
| Analysis | Genes Identified |
|---|---|
| Differentially Expressed Genes | 21,456 |
Gene Ontology enrichment revealed pathways associated with:
- Immune activation
- Extracellular matrix organization
- Cell adhesion
- Angiogenesis
- Tumor progression
Reactome pathway enrichment analysis was performed to identify biological processes and molecular mechanisms associated with differentially expressed genes.
Enriched pathways revealed involvement of:
- Immune system activation
- Antigen processing and presentation
- Cytokine signaling
- Extracellular matrix organization
- Cell adhesion
- Angiogenesis
- Cellular metabolism
- Cancer-associated signaling pathways
Cells were grouped into major biological compartments:
Tumor Cells
|
|
Immune Cells
|
|
Stromal Cells
The immune landscape revealed the contribution of:
- T cells
- NK cells
- Macrophages
- B cells
- Monocytes
Major cellular compartments were quantified to understand tumor ecosystem composition.
Example:
| Compartment | Cell Types |
|---|---|
| Immune | T cells, NK cells, B cells, Macrophages |
| Stromal | Fibroblasts, Endothelial cells |
| Tumor | Epithelial cells |
BreastCancer-scRNAseq/
├── data/
│
├── scripts/
│ ├── 01_Load_Data.R
│ ├── 02_QC.R
│ ├── 03_Filtering.R
│ ├── 04_Normalization.R
│ ├── 05_PCA.R
│ ├── 06_Clustering_UMAP.R
│ ├── 07_Marker_Genes.R
│ ├── 08_Cell_Annotation.R
│ ├── 09_DEG_Analysis.R
│ ├── 10_GO_KEGG.R
│ ├── 11_Reactome_Pathway.R
│ ├── 12_Marker_Heatmap.R
│ ├── 13_Marker_DotPlot.R
│ ├── 14_Cell_Composition.R
│ └── 15_TME_Analysis.R
│
├── figures/
│
├── results/
│
├── README.md
└── .gitignore
Install required packages:
install.packages("Seurat")
install.packages("tidyverse")
install.packages("patchwork")
BiocManager::install("SingleR")
BiocManager::install("celldex")
BiocManager::install("clusterProfiler")
BiocManager::install("ReactomePA")Run scripts sequentially:
source("scripts/01_Load_Data.R")
source("scripts/02_QC.R")
source("scripts/03_Filtering.R")
...
source("scripts/15_TME_Analysis.R")All results and figures will be generated automatically.
Key findings include:
- Identification of multiple immune and stromal populations.
- Characterization of the breast cancer tumor microenvironment.
- Cluster-specific marker genes associated with immune activation.
- Functional enrichment of pathways related to immune response, extracellular matrix organization, angiogenesis, and cell proliferation.
Potential future extensions include:
- CellChat analysis for cell-cell communication
- Monocle3 trajectory inference
- RNA velocity analysis
- Copy number variation inference
- Spatial transcriptomics integration
- Multi-omics integration
If this repository contributes to your research, please consider citing the original GSE176078 dataset and the Seurat framework.
Bano Rani
BS Bioinformatics
University of Agriculture Faisalabad
Research Interests:
- Single-cell Genomics
- Cancer Bioinformatics
- Machine Learning
- Computational Biology
- AI for Precision Medicine
If you find this project useful, consider giving it a ⭐ on GitHub.