Best way to gather gene and phenotype data for methylation sites?

Question

[Soso]

Hello!

I'm a total beginner in working with mehtylation data, wanting to interpret my findings.

I have a list of more than a hundred most significant (by p-value) methylation sites (as an example):

cg22674699, cg04104695, cg11308319, cg18081940, …

I would like to find the closest human gene to these, whether they are situated in an CpG island/shore and if they have been associated with a phenotype. I have scrolled through EWAS Atlas (https://bigd.big.ac.cn/ewas) and EWAS Data Hub (https://bigd.big.ac.cn/ewas/datahub/index) and I have found interesting information there, but would not want to gather all info manually.

Additionally I have looked at these IlluminaHumanMethylation annotation manifests (because of googling/biostaring my problem before sending my question here) for R but don't quite understand if they are what I'm looking for.

Warning, very possible newbie question coming! Is there a good way to gather gene and phenotype data for methylation sites? Some sites, programs or packages? Are package data really up to date? :)

Thank you in advance for all replies!

Answer 1

[Kevin Blighe]

The best source usually is the manifest file from the array manufacturer. These are usually distributed as TSV or CSV, which you can easily read into R.

Also take a look at the Bioconductor annotation packages for Illumina methylation arrays: https://www.bioconductor.org/packages/release/data/annotation/

These packages invariably just contain the manufacturer's data as R objects, though.

Are package data really up to date? :)

Yes and no... they may just contain the annotation provided by the manufacturer, so, consider that it may indeed be somewhat outdated.

Kevin

[Soso]

Thank you Kevin for a quick answer! I'll take a closer look at the Bioconductor site and keep learning for sure!

-Soso