Extracting gene sets from the Bioconductor reactome.db R package's SQLite backend with dplyr

reactome_dplyr.R

#' Retrieve Reactome sets of Entrez identifiers for a selected species
#'
#' @param species Scalar character, the species of interest, e.g. `Homo sapiens`
#' @importFrom dplyr tbl right_join select collect mutate check_dbplyr
#' @importFrom glue glue_sql glue
#' @importFrom checkmate assert_choice
#' @export
#' @return A named list of Entrez identifiers
#' @examples
#' ReactomeSets("Mycobacterium tuberculosis")
ReactomeSets <- function(
  species = c(
    "Bos taurus", "Caenorhabditis elegans", "Canis familiaris",
    "Danio rerio", "Dictyostelium discoideum",
    "Drosophila melanogaster", "Gallus gallus", "Homo sapiens",
    "Mus musculus", "Mycobacterium tuberculosis",
    "Plasmodium falciparum", "Rattus norvegicus",
    "Saccharomyces cerevisiae", "Schizosaccharomyces pombe",
    "Sus scrofa", "Xenopus tropicalis")) {
  species <- match.arg(species, verbose = FALSE)
  if (!requireNamespace("reactome.db")) {
    stop(paste("This function requires the suggested 'reactome.db'",
               "Bioconductor package. Please install it first."),
         call. = FALSE)
  }
  checkmate::assert_flag(verbose)
  dplyr::check_dbplyr()
  con <- reactome.db::reactome_dbconn()
  if (verbose == TRUE) {
    release <- dplyr::tbl(con, "metadata") %>%
      dplyr::filter(name == "DBSCHEMAVERSION") %>%
      dplyr::collect() %>%
      dplyr::pull("value")
    message(
      glue::glue("Retrieving gene sets from reactome release {release}.\n")
    )
  }
  dplyr::tbl(con, "pathway2name") %>%
    dplyr::filter(
      glue::glue_sql("path_name LIKE {paste0(species, ': %')}", .con = con)
    ) %>%
    dplyr::right_join(dplyr::tbl(con, "pathway2gene"), by = "DB_ID") %>%
    dplyr::select(gene_id, path_name) %>%
    dplyr::collect() %>%
    dplyr::mutate(path_name = sub(paste0(species, ': '), '', path_name)) %>%
    with(., split(gene_id, path_name))
}