Weighted Gene Co-expression Network Analysis (WGCNA)

Perform WGCNA analysis to identify co-expression modules and their eigenvalues. The function uses WGCNA package to construct weighted gene co-expression networks, detect modules, and calculate module membership and eigenvalues.

Usage

gly_wgcna(
  exp,
  powers = c(1:10, seq(12, 20, by = 2)),
  network_type = "unsigned",
  tom_type = "unsigned",
  min_module_size = 30,
  deep_split = 2,
  merge_cut_height = 0.25,
  add_info = TRUE,
  ...
)

gly_wgcna_(
  expr_mat,
  powers = c(1:10, seq(12, 20, by = 2)),
  network_type = "unsigned",
  tom_type = "unsigned",
  min_module_size = 30,
  deep_split = 2,
  merge_cut_height = 0.25,
  ...
)

Arguments

exp: A glyexp::experiment() object.
powers: A numeric vector of soft thresholding powers to test. Default is c(1:10, seq(12, 20, by = 2)).
network_type: Character string specifying network type. Allowed values are "unsigned", "signed", "signed hybrid". Default is "unsigned". Passed to the NetworkType`` argument of WGCNA::blockwiseModules()`.
tom_type: Character string specifying topological overlap matrix type. Allowed values are "none", "unsigned", "signed". Default is "unsigned". Passed to the TOMType`` argument of WGCNA::blockwiseModules()`.
min_module_size: Minimum module size for module detection. Default is 30. Passed to the minModuleSize`` argument of WGCNA::blockwiseModules()`.
deep_split: Integer value between 0 and 4. Provides a simplified control over how sensitive module detection should be to module splitting. Default is 2. Passed to the deepSplit`` argument of WGCNA::blockwiseModules()`.
merge_cut_height: Dendrogram cut height for merging of modules. Default is 0.25. Passed to the mergeCutHeight`` argument of WGCNA::blockwiseModules()`.
add_info: A logical value indicating whether to add variable information to the modules tibble. Only applicable to gly_wgcna().
...: Additional arguments passed to WGCNA::blockwiseModules().
expr_mat: A numeric matrix with variables as rows and samples as columns.

Value

A list with two elements:

tidy_result: A list containing two tibbles:
- modules: Module assignments and membership values containing the following columns:
  - variable: Variable name
  - module: Module assignment (color name)
  - membership: Module membership value (correlation with module eigengene)
- eigenvalues: Module eigenvalues containing the following columns:
  - module: Module name (color name)
  - sample: Sample name
  - eigenvalue: Module eigenvalue (first principal component of module expression)
raw_result: The raw WGCNA blockwiseModules object The list has classes glystats_wgcna_res and glystats_res.

Details

The function performs log2 transformation on the expression data (log2(x + 1)) before WGCNA analysis.

gly_wgcna() is the top-level API that works with glyexp::experiment() objects and supports the add_info parameter for joining experiment metadata.

gly_wgcna_() is the underlying API that works with matrices directly, providing more flexibility for users who don't use the glyexp package.

Analysis Steps:

Soft threshold selection using WGCNA::pickSoftThreshold()
Network construction and module detection using WGCNA::blockwiseModules()
Module membership calculation based on correlation with module eigengenes
Results formatting into standardized tibbles

Required packages

This function requires the following packages to be installed:

WGCNA` for weighted gene co-expression network analysis