Author: Claudia Costa
Thesis: A computational analysis to decipher different transcriptomic signatures regulating gliomas heterogeneity

Project Overview

This repository contains the computational data, enrichment results, and network topologies generated for this thesis. The study decodes the transcriptomic landscape regulating the dichotomy between the invasive (migraSi / "Go") and proliferative (migraNo / "Grow") phenotypes in Glioblastoma (GBM).

By integrating Differential Gene Expression (DGE) and Gene Set Enrichment Analysis (GSEA) with systems-level network inference, this workflow extracts highly specialized regulatory modules that capture the tumor's invasive machinery.

---

Repository Structure

The data is organized into four main directories corresponding to the sequential steps of the computational pipeline:

1. DGE/ (Differential Gene Expression)

Contains the baseline transcriptomic profiling comparing the migraSi and migraNo phenotypes.

• Normalized Counts: Raw RNA-Seq counts normalized utilizing the apeglm shrinkage estimator to handle dispersion and fold-change variance.
• Gene Lists: Complete lists of significantly up-regulated and down-regulated genes.
• Plots: Volcano plots and expression visualizations highlighting the phenotypic transcriptomic shift.

2. ORA/ (Over-Representation Analysis)

Contains the preliminary functional enrichment analyses.

• Background Set: The background set used to compute the enrichment p-values.
• Results: Tabular results of the enriched pathways and gene sets.

3. GSEA/ (Gene Set Enrichment Analysis)

Contains the advanced functional profiling utilizing the MSigDB collections.

• Complete Results: Full enrichment tables for the Hallmarks (H), Curated Pathways (C2), and Gene Ontology (C5) collections.
• Leading Edge Subsets: Isolated lists of the specific differential genes driving the enrichment scores within the core pathways.
• Plots: Enrichment score visualizations for the most important pathways and gene sets

4. GRNs/ (Gene Regulatory Networks)

Contains the topological network data used to identify context-specific regulatory modules.

• Extended Pathways: New regulatory pathways identified via the GSEA network extension.
• Filtered Network v2 (Edge Lists): The condition-specific topological edge lists for both the migraSi and migraNo architectures.

Note on Data Availability: > The complete, unpruned global Gene Regulatory Networks exceed GitHub's standard file size limits. Therefore, only the structurally Filtered Network v2 edge lists are hosted in this repository.

---

Reproducibility

The data provided in these directories directly corresponds to the figures, tables, and biological hypotheses presented in the Results chapters of the thesis.