get RefSeq mRNA mapping to human gene symbol via ensembl biomart

refseq_to_gene_symbol_biomart.R

# This script demonstrates how to download a mapping from RefSeq mRNA to human gene symbols by using the ensembl biomart service and the bioconductor `biomaRt` package in R.

source("https://bioconductor.org/biocLite.R")
biocLite("biomaRt")

library("biomaRt")

# work around bug in resolving host (https://support.bioconductor.org/p/74304/)
listMarts(host="www.ensembl.org")

# set the mart to use
ensembl <- useMart("ENSEMBL_MART_ENSEMBL", host="www.ensembl.org")

# find the dataset we want
datasets <- listDatasets(ensembl)
hg_row <- which(grepl('sapiens', datasets$dataset))
dataset <- as.character(datasets$dataset[hg_row])

# select the dataset as our source
ensembl <- useMart("ENSEMBL_MART_ENSEMBL", host="www.ensembl.org", dataset=dataset)

# extract the data
mapping <- getBM(attributes=c('refseq_mrna', 'external_gene_name'), mart=ensembl)

# clean it up (only rows with a RefSeq mrna id)
clean_mapping <- mapping[mapping$refseq != "",]

# size of the cleaned dataset
dim(clean_mapping)