enrichly

enrichly provides fast, tidy, local-first pathway enrichment analysis.

The package is designed for workflows that run many enrichment queries across many projects. It downloads Enrichr-style gene-set libraries once, loads them from local cache, and runs over-representation analysis without repeated API calls.

By default, enrichly_enrich() uses the standard Enrichr-compatible effective background size of 20,000 genes. This makes local p-values, odds ratios, and combined scores closely match enrichR::enrichr() for the same gene list and database while avoiding repeated web requests. Supply universe when you want an explicit project-specific background instead.

Install

remotes::install_github("oncologylab/enrichly")

Quick start

library(enrichly)

manifest <- enrichly_download()
db <- enrichly_load(manifest$path)

res <- enrichly_enrich(
  genes = c("TP53", "BAX", "CASP3"),
  db = db,
  query_id = "example"
)

enrichly_dotplot(res, "pathway_dotplot.pdf")

Default databases

enrichly defaults to the database set used by CraftGRN:

• GO_Biological_Process_2023
• GO_Cellular_Component_2023
• GO_Molecular_Function_2023
• Reactome_2022
• WikiPathways_2024_Human
• MSigDB_Hallmark_2020
• KEGG_2021_Human

Design

• Local-first enrichment after initial database download.
• Enrichr-compatible standard background and scoring by default.
• Tidy output tables.
• Result caching for repeated queries.
• Multicore support for high-throughput query batches.
• CraftGRN-compatible pathway dotplots.

Enrichr compatibility tests

The test suite includes frozen enrichR::enrichr() reference outputs for KEGG and MSigDB Hallmark queries. These offline tests compare local enrichly results against the real Enrichr API output for term coverage, overlap genes, p-values, adjusted p-values, odds ratios, combined scores, and ranking.