Hi,

I'd like to use annotationDbi to map various gene IDs (entrez,ensembl, etc) to official gene symbols quickly, but some of the gene symbols in the org.Mm.eg.db database aren't the official MGI gene symbols.  For example, mt-Cytb (entrez # 17711) has the gene symbol CYTB according to org.Mm.eg.db.

Am I using an inappropriate database?

Thanks,

Brendan

It seems that this is limited to the mitochondrial genome, and is due to NCBI's naming decisions, as explained here: org.Mm.eg.db gives wrong symbol for MT genes

The data for the org.Mm.eg.db package is parsed from files we download from NCBI. One of which is the gene_info.gz file. If I download that file and look for that Gene ID, this is what I get:

zcat gene_info.gz | awk '{if(NR == 1 || ($1 == 10090 && $2 == 17711)) print $2"\t"$3}'
GeneID  Symbol
17711   CYTB

And if I just look at the top 6:

zcat gene_info.gz | awk '{if(NR == 1 || ($1 == 10090)) print $2"\t"$3}' | head
GeneID  Symbol
11287   Pzp
11298   Aanat
11302   Aatk
11303   Abca1
11304   Abca4
11305   Abca2
11306   Abcb7
11307   Abcg1
11308   Abi1

We are simply repackaging the data we can get from NCBI, and if the files they supply say the gene symbol is CYTB, that's what will be in the org.Mm.eg.db package.