A.V. AKILA RAVIHANSA PERERA – A0212216X

TRAN KHANH HUNG – A0212253W

• WGCNA

Weighted gene correlation network analysis (WGCNA) is a powerful method that uses a topological overlap module approach for constructing co-expression networks based on gene expression data. This method involves reconstructing gene co-expression modules and summarizing modules using module eigengenes (ME) and intramodular hub genes.

• Gene Enrichment and Pathway Analysis

Biologically interesting modules were identified using Fisher's exact test. The overlapping and union sets of genes from theses interesting gene module pairs were subjected to Gene Set Enrichment Analysis using topGO package).

Three gene expression datasets for three cancer types (GBM, OV, BRCA) were selected from TCGA (The Cancer Genome Atlas).

• Glioblastoma Multiforme (GBM) gene expression by RNAseq https://tcga.xenahubs.net/download/TCGA.GBM.sampleMap/HiSeqV2_PANCAN.gz

• Ovarian Serous Cystadenocarcinoma (OV) gene expression by RNAseq https://tcga.xenahubs.net/download/TCGA.OV.sampleMap/HiSeqV2_PANCAN.gz

• Breast Invasive Carcinoma (BRCA) gene expression by RNAseq https://tcga.xenahubs.net/download/TCGA.BRCA.sampleMap/HiSeqV2_PANCAN.gz

Description	BRCA	GBM	OV
Filter threshold (CV)	0.5	0.5	0.5
Sample Clustering Dendrogram
deepSplit	3	3	3
minClusterSize	30	30	30
Number of gene modules	19	19	20

Scale Free Topology Model	Mean Connectivity	Selected Soft Threshold
		9
		3
		6

Genes	Module Eigengenes

Network Heatmap	Eigengene Adjacency Heatmap	Eigengene Dendrogram

BRCA and GBM	GBM and OV	OV and BRCA

• Download and extract datasets to ./data directory

• Install dependencies (R packages)

  • Run 0_install_dependencies.R

• Gene filtering

  • Run 1_wgcna_cluster.R

• Build gene expression network and identify gene modules

  • Run 2_module_detection.R

• Generate gene network plots

  • Run 3_network_visualization.R

• Gene Enrichment Analysis

  • Run 4_gene_enrichment.R