This method uses all_genes and contains >1,200 more genes
than hgnc2entrez due to querying a more comprehensive
and frequently updated database.
data("hgnc2entrez_orthogene")An object of class data.frame with 25721 rows and 2 columns.
The code to prepare the .Rda file file from the marker file is:
#### OLD METHOD: biomaRt ####
library("biomaRt")
human <- useMart(host="www.ensembl.org",
"ENSEMBL_MART_ENSEMBL",
dataset="hsapiens_gene_ensembl")
attrib_hum = listAttributes(human)
hgnc_symbols = getBM(attributes=c("hgnc_symbol","entrezgene"), mart=human)
colnames(hgnc_symbols) = c("hgnc_symbol","entrez")
hgnc_symbols = hgnc_symbols[hgnc_symbols$hgnc_symbol!=""]
hgnc_symbols = hgnc_symbols[!is.na(hgnc_symbols$entrez),]
hgnc2entrez = hgnc_symbols
usethis::use_data(hgnc2entrez,overwrite = TRUE)#### NEW METHOD: orthogene ####
library(MAGMA.Celltyping); library(orthogene); library(dplyr);
gene_map <- orthogene::all_genes(species = "human",
method = "gprofiler",
target = "ENTREZGENE_ACC",
ensure_filter_nas = FALSE)
hgnc2entrez_orthogene <- gene_map |>
dplyr::select(hgnc_symbol = Gene.Symbol,
entrez = target) |>
unique()
#### Compare to other dataset ####
dt1 <- hgnc2entrez |> dplyr::filter(!hgnc_symbol
!entrez
unique()
dt2 <- hgnc2entrez_orthogene |> dplyr::filter(!hgnc_symbol
!entrez
unique()
message("hgnc2entrez_orthogene has ",
formatC(nrow(dt2) - nrow(dt1), big.mark = ","),
" more genes than original method.")
usethis::use_data(hgnc2entrez_orthogene, overwrite = TRUE)