Systems-Level Host Defense Analysis Pipeline

This repository contains the core analytical code used to identify "Rewired" proteins and functional defense modules in Fusarium species. The scripts provided here are simplified, non-looping versions intended to demonstrate the underlying methodology logic. They are designed to be adaptable to any user-provided RNA-seq and Interactome dataset.

Methodology Reference

The Differential Interactome methodology used to calculate protein-protein interaction rewiring is based on the "RelA" approach described in:

Gulfidan, G., Turanli, B., Beklen, H. et al. (2020). "Pan-cancer mapping of differential protein-protein interactions". Scientific Reports 10, 3272. https://doi.org/10.1038/s41598-020-60127-x

Repository Structure

The workflow is divided into three logical phases:

1. 01_differential_interactome.R

   • Input: Normalized Gene Expression Matrix (TPM/FPKM) and a Reference Interactome (PPI).
   • Process: Calculates Pearson Correlation Coefficients (PCC) for all interactions in Control vs. Treatment conditions. Defines "Rewiring" as the absolute difference in correlation ($\Delta PCC$).
   • Output: A list of proteins ranked by their "Differential Degree" ($k_{diff}$ — the sum of rewired edges).

2. 02_rewired_classification.R

   • Input: Output from Step 01 and a list of Differentially Expressed Genes (DEGs) from standard tools (e.g., DESeq2).
   • Classification:
     • Rewired Proteins: Top 25% highest Differential Degree AND Stable Expression ($|logFC| < 1$).
     • Transcriptional Targets: Significant DEGs ($P_{adj} < 0.05$).
   • Goal: To distinguish proteins undergoing conformational/functional changes (Rewiring) from those undergoing abundance changes (Expression).

3. 03_module_discovery.R

   • Input: The sub-network of identified Rewired Proteins.
   • Process: Applies Louvain Modularity detection to identify dense functional clusters (modules) within the rewired network.
   • Output: List of functional modules representative of the system's coordinated defense response.

Usage

These scripts use placeholders (e.g., "path/to/expression_matrix.csv") and should be updated with your specific file paths.

# Example: Run the differential interactome calculation
source("01_differential_interactome.R")

Requirements

• R (v4.0+)
• igraph
• dplyr
• WGCNA (optional, for fast correlation) or base R