What it does

This repo contains scripts and data files that find Gencode gene IDs with given gene names. It relies on 3 resources to find the hits, and the search is attempted in the following priority order:

• Gencode GTF: https://www.gencodegenes.org/human/ (currently using v34)
• NCBI gene records: https://www.ncbi.nlm.nih.gov/gene/?term=homo+sapiens (downloaded in Feb 2020)
• HGNC gene aliases: https://www.genenames.org/download/custom/ (downloaded in June 2020)

Note: if a gene name is found to be linked with more than one Gencode ID it will be reported as no hit.

Load function and data in R

library(RCurl)
script=getURL("https://raw.githubusercontent.com/weishwu/find_gencodeID/master/find_gencodeID.r", ssl.verifypeer = FALSE)
load(url("https://raw.githubusercontent.com/weishwu/find_gencodeID/master/queryTabs.RData"))
eval(parse(text = script))

Example run

• Read input from a file name which has to end with ".txt". The file contains gene names in the first column, one name per line.

hits=find_gencodeID("ENSG_Missing.txt")

• Read input from the variable name of a vector of characters.

hits=find_gencodeID(ENSG_Missing)

Read below only if data needs to be re-created

• Merge NCBI and HGNC records with Gencode GTF to find the unique matches (each NCBI/HGNC record can only be linked with one Gencode ID; each gene name can only be linked with on Gencode ID).
• Requried inputs: gencode.v34.annotation.gtf.zip, NCBI_genes.txt.zip, HGNC_gene_aliases.txt.zip (downloaded from the links listed above)

python gencodeID_queryTables.py gencode.v34.annotation.gtf.zip

• Combine the gene name and ID tables created above into an R object

Rscript gencodeID_queryTables.r