Kruskal-Wallis test for Differential Expression Analysis

Perform Kruskal-Wallis test for glycomics or glycoproteomics data. The function supports non-parametric comparison of multiple groups. For significant results, Dunn's post-hoc test is automatically performed. P-values are adjusted for multiple testing using the method specified by p_adj_method.

Usage

gly_kruskal(
  exp,
  group_col = "group",
  p_adj_method = "BH",
  add_info = TRUE,
  ...
)

gly_kruskal_(expr_mat, groups, p_adj_method = "BH", ...)

Arguments

exp: A glyexp::experiment() object containing expression matrix and sample information.
group_col: A character string specifying the column name of the grouping variable in the sample information. Default is "group".
p_adj_method: A character string specifying the method to adjust p-values. See p.adjust.methods for available methods. Default is "BH". If NULL, no adjustment is performed.
add_info: A logical value. If TRUE (default), variable information from the experiment will be added to the result tibble. If FALSE, only the statistical results are returned. Only applicable to gly_kruskal().
...: Additional arguments passed to stats::kruskal.test().
expr_mat: A numeric matrix with variables as rows and samples as columns.
groups: A factor or character vector specifying group membership for each sample. Must have at least 2 levels. Character vectors will be automatically converted to factors.

Value

A list containing two elements:

tidy_result: A list containing:
- main_test: A tibble with Kruskal-Wallis test results containing the following columns:
  - variable: Variable name
  - statistic: Kruskal-Wallis test statistic
  - p_value: Raw p-value from Kruskal-Wallis test
  - parameter: Degrees of freedom
  - method: Statistical method used
  - p_adj: Adjusted p-value (if p_adj_method is not NULL)
  - post_hoc: Significant group pairs from post-hoc test
- post_hoc_test: A tibble with pairwise comparison results containing the following columns:
  - variable: Variable name
  - group1: First group in comparison
  - group2: Second group in comparison
  - p_value: Raw p-value from Dunn's test
  - p_adj: Adjusted p-value from Dunn's test
raw_result: A list containing:
- main_test: A list of raw kruskal.test objects.
- post_hoc_test: A list of raw dunnTest objects.

Details

The function performs log2 transformation on the expression data (log2(x + 1)) before statistical testing. At least 2 groups are required in the grouping variable.

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

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

Post-hoc Test: Dunn's test with Holm correction for multiple comparisons (FSA::dunnTest()) is performed for variables with significant main effects (p_adj < 0.05).

Required packages

This function requires the FSA package for Dunn's post-hoc test.