Systematic review of package repositories for the R language (CRAN, GitHub, Bioconductor)
Main Article Content
Abstract
R packages are mainly distributed through three repositories: CRAN, Bioconductor and GitHub. Although CRAN and Bioconductor offer thematic classification systems (CRAN Task Views and BiocViews), no tool allows systematic searches within package metadata, particularly on GitHub, where much of the software development takes place. In this note, we present three R functions that share a common search engine and enable systematic queries of CRAN, Bioconductor and GitHub. Each function validates the query, normalises it to support Boolean operators (and, or), aggregates the descriptive information of each package into a common text, filters by exact term matching and returns a structured table of results. Thus, the same query can be used across all three repositories without knowing the specific search syntax of each one. As an example, a query on sequence alignment and workflows identified 6 packages in CRAN (out of 23 749), 6 in Bioconductor (out of 2361) and 575 repositories on GitHub. These results are used to illustrate further analyses, such as the number of packages published per year and their downloads, package dependencies represented as a graph, and the relationship between repository creation date and number of stars. These functions, whose code is available in a public repository, facilitate both the exploration of existing tools before developing a new package and the systematic review of the R package ecosystem.
Downloads
Article Details

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Accepted 2026-09-29
Published 2026-11-10