A Protocol for Weighted Gene Co-expression Network Analysis With Module Preservation and Functional Enrichment Analysis for Tumor and Normal Transcriptomic Data

Equipment

1. Personal computer with Windows, macOS, or a Unix-based operating system

Note: For these analyses, we recommend a computer with at least 32 GB of RAM and a modern multi-core processor.

Software and datasets

1. R software environment (> 4.4.0) (https://www.r-project.org/)

2. RStudio integrated development environment (> 1.4.0) (https://rstudio.com/)

3. WGCNA_1.73 (https://cran.r-project.org/web/packages/WGCNA/index.html)

4. DESeq2_1.46.0 (https://bioconductor.org/packages/release/bioc/html/DESeq2.html)

5. genefilter_1.88.0 [22] (https://www.bioconductor.org/packages/release/bioc/html/genefilter.html)

6. tidyverse_2.0.0 (https://cran.r-project.org/web/packages/tidyverse/index.html)

7. dendextend_1.19.0 (https://cran.r-project.org/web/packages/dendextend/index.html)

8. gplots_3.2.0 (https://cran.r-project.org/web/packages/gplots/index.html)

9. clusterProfiler_4.14.4 [23] (https://bioconductor.org/packages/release/bioc/html/clusterProfiler.html)

10. org.Hs.eg.db_3.20.0 (https://bioconductor.org/packages/release/data/annotation/html/org.Hs.eg.db.html). All gene annotations were performed using the org.Hs.eg.db package (version 3.20.0), which is consistent with the GRCh38 human genome assembly of the TCGA data

11. ggplot2_3.5.1 (https://cran.r-project.org/web/packages/ggplot2/index.html)

12. ggpubr_0.6.0 (https://cran.r-project.org/web/packages/ggpubr/index.html)

13. VennDiagram_1.7.3 ( https://cran.r-project.org/web/packages/VennDiagram/index.html)

14. dplyr_1.1.4 (https://cran.r-project.org/web/packages/dplyr/index.html)

15. GO.db_3.20.0 (https://bioconductor.org/packages/release/data/annotation/html/GO.db.html)

16. Code for WGCNA and preservation module analysis can be accessed through this GitHub link: https://github.com/biocoms/WGCNA

17. Dataset

Data can be directly downloaded through the TCGA portal. The analysis used gene expression data from the GDC Data Release v38.0, downloaded on October 18, 2023. The dataset is publicly available as a compressed file (GeneExpression.zip) in the GitHub repository: https://github.com/biocoms/WGCNA

if (!require("BiocManager", quietly = TRUE))

        install.packages("BiocManager")

   BiocManager::install("TCGAbiolinks")

TCGAbiolinks:::getProjectSummary("TCGA-HNSC")

ge_query <- GDCquery( project = "TCGA-HNSC",

                data.category = "Transcriptome Profiling",

                data.type = "Gene Expression Quantification")

GDCdownload(ge_query)

output_m_query <- getResults(ge_query)

normal_sample <- output_m_query$sample.submitter_id[output_m_query$sample_type == "Solid Tissue Normal"]

tumor_sample <- output_m_query$sample.submitter_id[output_m_query$sample_type == "Primary Tumor"]

ge_data <-GDCprepare(ge_query, summarizedExperiment = TRUE)

# Extracts the expression values

expression_matrix <- assay(ge_data)

expression_df <- as.data.frame(expression_matrix)

expression_df <- cbind(Gene = rownames(expression_matrix), expression_df)

colnames(expression_data) <- gsub("-.*$", "", colnames(expression_data))

# Remove extra parts

# Subset the expression data to include only normal samples/tumor sample

normal_expression_data <- expression_data[, colnames(expression_data) %in% normal_samples]

tumor_expression_data <- expression_data[, colnames(expression_data) %in% tumor_samples]

write.csv(normal_expression_data, “OSCC_TCGA_gene_expression_normal.csv)

write.csv(tumor_expression_data, “OSCC_TCGA_gene_expression_tumor.csv)

Based on the different purposes of research, we can manipulate the dataset by removing or filtering out some samples.

Caution: Troubleshooting and environment setup: this protocol was developed and tested using R version 4.4.0 or higher. However, we recognize that users may have different R versions. Here, we provide solutions for various compatibility issues:

For R versions 4.2.x–4.3.x:

Most functions should work, but some package versions may differ. If you encounter issues:

1. For WGCNA installation:

For older R versions, use the archived WGCNA version

install.packages("https://cran.r-project.org/src/contrib/Archive/WGCNA/WGCNA_1.72-1.tar.gz", repos = NULL, type = "source")

2. For Bioconductor packages:

if (!requireNamespace("BiocManager", quietly = TRUE))

install.packages("BiocManager")

BiocManager::install(version = "3.17") # For R 4.3.x

BiocManager::install(version = "3.16") # For R 4.2.x

For R 4.2.0 (Windows): Due to package compatibility issues with the older R version, users may encounter installation problems. Follow these steps:

1. Clean installation approach:

Remove any partially installed packages

remove.packages(c("WGCNA", "htmlTable", "knitr", "xfun"))

Clear package cache

unlink(.libPaths()[1], recursive = TRUE)

dir.create(.libPaths()[1])

2. Use the compatibility script (recommended):

GitHub link: https://github.com/biocoms/WGCNA/blob/main/R_4.2.0_compatibility_script.R

3. Known issues and workarounds:

If htmlTable fails, install from source with install.packages("htmlTable", type = "source")

If the WGCNA function is not found, restart R after installation and explicitly load with library(WGCNA)

For quotation mark issues, ensure you are using straight quotes, not curly quotes.

For R versions < 4.2:

We strongly recommend upgrading to R 4.2 or higher. If an upgrade is not possible, use WGCNA version 1.70.3 or earlier. Some visualization features may not be available.

Procedure