Breast Cancer Exploratory Data Analysis in R

Exploratory analysis of breast invasive carcinoma (BRCA) clinical data using R and RStudio. This project was developed as part of the Research Methodology I module (Unit 1, Activity 1) for the Master’s in Bioinformatics.

---

Overview

This project performs an exploratory analysis of The Cancer Genome Atlas (TCGA) breast cancer clinical data. The analysis includes:

• Data ingestion & cleaning: Loading and handling missing values from the TCGA clinical patient file.
• Feature engineering: Grouping AJCC tumor stages into broader categories (Stage I–IV).
• Descriptive statistics: Boxplots and mean diagnosis age by tumor stage.
• Inferential statistics: Normality testing (Shapiro-Wilk), homoscedasticity testing (Levene), and comparison of means between Stage I and Stage III (Wilcoxon and t-tests).
• Survival analysis: Kaplan–Meier survival curves stratified by progesterone receptor (PR) status.

---

Repository Structure

File	Description
analisis_exploratorio_JL.R	Main R script containing the full analysis pipeline
nationwidechildrens.org_clinical_patient_brca.txt	Raw TCGA-BRCA clinical patient data (tab-separated)
Analisis_exploratorio_JavierLuque.pdf	Final report with results and interpretation (PDF)
0JCP001104_UA1_AA1.pdf	Assignment guidelines and instructions (PDF)
boxplot_edad.png	Output figure: boxplot of diagnosis age by tumor stage
survival_plot_PR.png	Output figure: Kaplan–Meier survival plot by PR status
breast_data_stage.csv	Processed CSV dataset with diagnosis age and grouped stages (generated by the script)
README.md	This file

---

Dependencies

The following R packages are required to run the analysis:

• readr
• dplyr
• car
• survival
• survminer

You can install the missing packages with:

install.packages(c("readr", "dplyr", "car", "survival", "survminer"))

Note: This project was developed and tested on R with RStudio.

---

How to Run

1. Clone the repository:

   git clone https://github.com/<your-username>/<your-repo-name>.git
   cd <your-repo-name>

2. Open analisis_exploratorio_JL.R in RStudio (or your preferred R environment).

3. Update the file paths inside the script if necessary (the script currently uses absolute paths pointing to the project directory).

4. Run the script line-by-line or source the entire file to reproduce the analysis, figures, and the breast_data_stage.csv output.

---

Key Results

• Age vs Stage: The boxplot analysis reveals differences in median diagnosis age across tumor stages.
• Statistical Comparison: Between Stage I and Stage III, normality and variance assumptions were checked, followed by non-parametric and parametric mean-comparison tests.
• Survival Analysis: The Kaplan–Meier curves show survival probabilities over time stratified by PR status, with log-rank p-values and risk tables included.

Detailed interpretation and conclusions are available in the accompanying PDF report (Analisis_exploratorio_JavierLuque.pdf).

---

Author

Javier Luque Serrano
Master’s in Bioinformatics European University of Madrid Research Methodology I – Module 4, Unit 1, Activity 1

---

License

This project is licensed under the MIT License.

The raw clinical data originates from The Cancer Genome Atlas (TCGA) and is publicly available through the GDC Data Portal.