CRPC WGCNA Metastasis 2025

This repository contains the reproducible R code used in the manuscript:

(https://doi.org/10.1016/j.adcanc.2025.100132)

The project aims to:

• Download and preprocess the public microarray dataset GSE74685 (castration-resistant prostate cancer metastases).
• Perform differential expression analysis (Bone vs Visceral metastatic sites).
• Build a WGCNA network and detect co-expression modules.
• Correlate modules with metastatic site traits.
• Extract long non-coding RNA (lncRNA) candidates (TP53TG1, RFPL1S, DLEU1).
• Export gene–module assignments, module–trait correlations, candidate lists, and example figures.

All analyses are implemented in R and organized as a small, self-contained pipeline.

---

Directory Structure

crpc-wgcna-metastasis-2025.Rproj – RStudio project file (recommended entry point).

scripts/

Main analysis scripts:

• Script 01_load_packages.R – Installs and loads packages (GEOquery, Biobase, limma, WGCNA, dynamicTreeCut, tidyverse), sets global options.
• Script 02 - download & preprocess GSE74685.R – Downloads GSE74685, filters, and saves processed data.
• Script 03 - DE analysis (Bone vs Visceral).R – Performs differential expression analysis with limma.
• Script 04 - WGCNA network and modules (GSE74685).R – Builds WGCNA network, detects modules, correlates with traits, exports results.
• Script 05 - Extract module genes & lncRNA candidates (GSE74685).R – Generates candidate gene and lncRNA lists combining DE and WGCNA results.
• Optional GPL annotation helper script – Builds clean annotation tables and identifies lncRNA biotypes.

data/

Processed expression matrices, phenotype tables, and WGCNA objects. Files are regenerable and excluded from version control.

results/

CSV outputs from DE and WGCNA analyses:

• DEG tables
• Module–trait correlation tables
• Gene/module/lncRNA candidate lists

figures/

Key plots from the pipeline:

• Sample clustering dendrograms
• Scale-free topology plots
• Gene dendrogram with module colours
• TOM and module–trait heatmaps

docs/

Lightweight documentation, such as platform/GPL annotation files.

---

Requirements

• R ≥ 4.2
• Packages: GEOquery, Biobase, limma, WGCNA, dynamicTreeCut, tidyverse

All required packages are loaded (and installed if missing) by scripts/Script 01_load_packages.R.

---

Running the Pipeline

1. Clone or download the repository.

2. Open crpc-wgcna-metastasis-2025.Rproj in RStudio.

3. Run scripts in order:

   1. 01_load_packages.R
   2. 02 - download & preprocess GSE74685.R
   3. 03 - DE analysis (Bone vs Visceral).R
   4. 04 - WGCNA network and modules (GSE74685).R
   5. 05 - Extract module genes & lncRNA candidates (GSE74685).R

4. Outputs will appear in data/, results/, and figures/.

5. Figures can be directly used in manuscripts or adapted for new analyses.

---

License

Code is released under the MIT License. Underlying expression data (GSE74685) remain subject to original GEO/authors’ terms.

---

Author

Dr. Roozbeh Heidarzadehpilehrood Human Geneticist – Transcriptomics & ncRNA biomarkers Email: heidarzadeh.roozbeh@gmail.com

---

Citation

If using this code or parts of the pipeline, cite both:

1. Mehrabi T, Heidarzadeh R, et al. Dysregulated key long non-coding RNAs TP53TG1, RFPL1S, DLEU1 in prostate cancer. Advances in Cancer Biology – Metastasis. https://doi.org/10.1016/j.adcanc.2025.100132

2. Heidarzadeh R. (2025). heidarzadehroozbeh-cmyk/crpc-wgcna-metastasis-2025: First public release – CRPC WGCNA metastasis pipeline (v1.0.0). https://doi.org/10.5281/zenodo.17830034

---