🧬 Gene Expression Analysis in Cancer using RNA-Seq

Author: Pranava Upparlapalli
Date: June 2025

---

📘 Project Summary

This project analyzes RNA-Seq gene expression data across five major cancer types:

• BRCA – Breast invasive carcinoma
• KIRC – Kidney renal clear cell carcinoma
• LUAD – Lung adenocarcinoma
• COAD – Colon adenocarcinoma
• PRAD – Prostate adenocarcinoma

Using the DESeq2 pipeline in R, we performed differential expression analysis, PCA, clustering, and visualized key patterns to identify cancer-specific gene signatures.

---

📁 Project Structure

CANCER-RNA-SEQ-ANALYSIS/
│
├── src.R # Main code
├── plots/                     # All generated plots
│   ├── heatmap_expression.png
│   ├── ma_plot.png
│   ├── sample_distribution.png
│   ├── silhouette_plot.png
│   └── volcano_plot.png
│
├── Notebook/                  # RMarkdown and Word version of the analysis
│   ├── Project V2.Rmd
|
├── index.html                 # HTML output of the analysis
├── README.md                  # This file
└── LICENSE                    # License file

---

📦 Data Availability

The RNA-Seq count data used in this project was sourced from Kaggle. Due to its large size, the data files (data.csv and labels.csv) are not included in this GitHub repository.

You can download the dataset from Kaggle at:

🔗 Kaggle Dataset – RNA-Seq Expression Data for Cancer Types

After downloading, place the files inside the data/ directory as follows:

CANCER-RNA-SEQ-ANALYSIS/
└── data/
    ├── data.csv
    └── labels.csv

---

🧪 Methods Overview

1. Data Preprocessing

   • Merged data.csv and labels.csv.
   • Removed missing values and ensured valid class labels.

2. Differential Expression (DESeq2)

   • Used DESeq2 to normalize, model, and identify significantly differentially expressed genes (padj < 0.05).

3. Visualization

   • Volcano plot and MA plot to show fold-change vs significance.
   • PCA to visualize clustering of samples.
   • Heatmap of top 50 most variable genes.
   • K-means clustering and silhouette plot for unsupervised pattern discovery.

---

📊 Visualizations

🔍 Sample Distribution

Shows the distribution of samples across cancer types.

---

🌋 Volcano Plot

Highlights significantly up/down-regulated genes.

---

🧮 MA Plot

Visualizes mean expression vs. fold change.

---

🔬 Silhouette Plot

Assesses clustering performance using PCA + K-means.

---

🔥 Heatmap of Top 50 Variable Genes

Displays patterns of gene expression across cancer types.

---

🧠 Key Findings

• Thousands of genes are differentially expressed between cancers.
• PCA showed distinct clustering of cancer types.
• K-means clustering aligned well with known classes.
• Some cancer types share expression signatures, indicating potential shared biology.

---

⚙️ Technologies Used

• R (v4.3.2)
• DESeq2
• ggplot2
• pheatmap
• factoextra
• tidyverse

---

📝 How to Run

1. Clone the repository:

   git clone https://github.com/yourusername/CANCER-RNA-SEQ-ANALYSIS.git
   cd CANCER-RNA-SEQ-ANALYSIS

2. Download the data from Kaggle and place it under data/.

3. Open Project V2.Rmd in RStudio.

4. Knit to HTML to generate index.html.

---

📃 License

This project is licensed under the MIT License.
See LICENSE for details.

---

🙋‍♂️ Contact

For questions, feedback, or collaborations:
📧 pxu.bioinfo@gmail.com
🔗 LinkedIn
🌐 Portfolio

---