enrichit: C++ Implementations of Functional Enrichment Analysis

The enrichit package provides fast, efficient, and lightweight implementations of common functional enrichment analysis methods, including Over-Representation Analysis (ORA) and Gene Set Enrichment Analysis (GSEA). The core algorithms are implemented in C++ using Rcpp to ensure high performance, making it suitable for analyzing large datasets or running simulations.

Installation

You can install the development version of enrichit from GitHub using devtools:

# install.packages("devtools")
devtools::install_github("YuLab-SMU/enrichit")

Features

• High Performance: Core calculations are written in C++.
• ORA: Standard hypergeometric test for over-representation analysis.
• GSEA:
  • Multilevel: Efficient p-value estimation for high-significance results (similar to fgsea).
  • Permutation: Standard permutation-based p-value calculation.
  • Adaptive: Adaptive permutation approach.
• GSON Support: Native support for GSON objects for gene set management.
• Standardized Output: Returns enrichResult and gseaResult objects compatible with the clusterProfiler ecosystem.

Usage

Over-Representation Analysis (ORA)

library(enrichit)

# Example gene sets
gene_sets <- list(
  pathway1 = paste0("Gene", 1:50),
  pathway2 = paste0("Gene", 51:100)
)

# Define a universe and a list of significant genes
universe <- paste0("Gene", 1:1000)
sig_genes <- paste0("Gene", 1:20) # Significant genes

# Run ORA
ora_res <- ora(gene = sig_genes, 
               gene_sets = gene_sets, 
               universe = universe)

print(ora_res)

Gene Set Enrichment Analysis (GSEA)

library(enrichit)

# Generate a ranked gene list
set.seed(123)
geneList <- sort(rnorm(1000), decreasing = TRUE)
names(geneList) <- paste0("Gene", 1:1000)

# Define gene sets
gene_sets <- list(
  pathway1 = paste0("Gene", 1:50),  # Enriched at top
  pathway2 = paste0("Gene", 951:1000) # Enriched at bottom
)

# Run GSEA
gsea_res <- gsea(geneList = geneList, 
                 gene_sets = gene_sets, 
                 method = "multilevel")

print(gsea_res)