Question: How to convert Illumina HumanMethylation27 BeadChip ID's to gene names?
0
17 months ago by
arbet00310
arbet00310 wrote:

I have DNA Methylation data from GSE26126 (you can access the data here) which uses the Illumina HumanMethylation27 BeadChip technology.  Here is an example of what some of the gene IDs look like:

"cg00027083" "cg00035347" "cg00446235" "cg00563845" "cg00955230"

How do I convert these ID's into the standard gene names/symbols in R?

Thanks!

modified 17 months ago by James W. MacDonald51k • written 17 months ago by arbet00310
1
17 months ago by
United States
James W. MacDonald51k wrote:

First, you should note that the mapping of CpGs to genes is not a straightforward thing to do, so converting CpG IDs to gene symbols isn't the same thing as e.g., converting an expression array probe ID to the gene it measures. The FDb.InfiniumMethylation.hg19 package is useful for doing the mapping you want. For an example, see the help for getNearest:

Annotate a group of probes supplied as a GenomicRanges object

Description:

This function encapsulates the common task of finding what
gene/tx/tss is closest to a list of probes, as might be the case
after a model fit.

Author(s):

Tim Triche, Jr.

Examples:

## not run
##
## hm450 <- get450k()
## probenames <- c("cg16392865", "cg00395291", "cg09310185", "cg21749424")
## probes <- hm450[probenames]
##
## getNearestTSS(probes)
## getNearestTranscript(probes)
##
## Notice that cg16392865 sits within PTCRA, but the nearest TSS is CNPY3.
##

But also note that you would first use get27k, not get450k.