Analysis of O-GlcNAcylation Dysregulation in Cancer

This repository contains Jupyter notebooks to investigate the impact of O-GlcNAcylation dysregulation on cancer through data-driven methodologies. The analysis uses gene expression data from TCGA (The Cancer Genome Atlas) to quantify dysregulation.

---

Files in This Repository

1. 1_data_download.ipynb
   This notebook provides step-by-step instructions to download RNA-seq gene expression data from the GDC portal using the GDC Data Transfer Tool. It specifies how to structure and organize downloaded files for further analysis.

2. 2_KDE_generation.ipynb
   This notebook generates Kernel Density Estimates (KDEs) for visualizing and analyzing OGT and OGA expression distributions across different cancer types.

3. 3_simulation_runs.ipynb
   This notebook simulates data to compare different metrics, including KDE-based measures and other regulation measures, for distinguishing between healthy and cancerous tissues.

4. 4_modeling.ipynb
   This notebook performs modeling using real cancer datasets, quantifying the relationship between OGT and OGA expression. It provides statistical insights into O-GlcNAcylation dysregulation across cancers.

---

Prerequisites

• Python 3.8 or later
• Jupyter Notebook
• Required Python libraries: numpy, pandas, matplotlib, seaborn, scipy, sklearn (refer to each notebook for specific imports)

---

Data Requirements and Organization

TCGA Gene Expression Data

For the analysis, RNA-seq gene expression data must be downloaded and organized as follows:

1. Go to the GDC Portal.

2. Select the cancer type of interest by tissue (e.g., Breast, Blood and Bone Marrow, etc).

3. Under the Repository, apply the following filters:

   • Experimental Strategy: RNA-seq
   • Data Category: Transcriptome Profiling
   • Data Type: Gene Expression Quantification

4. Use the GDC Data Transfer Tool to download the sample gene expression data.

5. Organize the downloaded .tsv files in the following structure:

   /data/TCGA_GeneExpression/{cancer}/gene_expression/
   

   Replace {cancer} with the specific cancer type.

Example Directory Structure

data/
└── TCGA_GeneExpression/
    ├── Kidney/
    │   └── gene_expression/
    │       ├── sample_1.tsv
    │       ├── sample_2.tsv
    │       └── ...
    ├── Lung/
    │   └── gene_expression/
    │       ├── sample_1.tsv
    │       ├── sample_2.tsv
    │       └── ...

---

Usage Instructions

1. Data Download:
   Run the 1_data_download.ipynb notebook to confirm data requirements and download the necessary files.

2. KDE Generation:
   Execute 2_KDE_generation.ipynb to compute and visualize KDEs for the selected cancer types.

3. Simulation Runs:
   Use 3_simulation_runs.ipynb to simulate data and compare different measures of regulation.

4. Modeling:
   Run 4_modeling.ipynb for application of the methodology to TCGA datasets.

---

Notes

• Ensure the data is preprocessed as described in 1_data_download.ipynb before proceeding to the analysis notebooks.
• Modify paths and cancer types in each notebook as necessary.

---