Question Posted 01/06/17:
Hi folks,

I'm working with ADNI GWAS data, and I've discovered that these files use Illumina's kgp* format for naming SNPs rather than the rs* format used by NCBI. I'd like to find the rs ids for these SNPs so that I can integrate the ADNI GWAS data with other data using rs ids. I see that this question has been asked before (e.g., here:, and the recommended solution is to use the UCSC table browser, which limits queries to 1000 SNPs. Since ADNI's Omni 2.5 GWAS data includes more than 2 million SNPs, this seems like an impractical solution.

Some googling reveals posts from other researchers (using non-ADNI data) who have been advised to contact Illumina directly for a mapping between kpg and rs identifiers. In fact, Illumina's website gives a number of reference files for their many different arrays (see, but I'm not sure which pertains to ADNI's GWAS data. There are a number of file sets labeled Omni 2.5, with dates around the spring of 2013 or soon after. I'd appreciate your help in figuring out which of these is the best reference for the ADNI data--alternatively, I'd also be grateful if you can point me to another method for making this conversion which does not involve repeated lookups to the UCSC table browser.


Jeff Phillips
Penn FTD Center
University of Pennsylvania
Response posted 01/06/17 by Kwangsik Nho:
Usually you can use two ways. 1. contact and consult Illumina directly (you can find the ADNI genotyping information in the paper (
2. download SNP annotation files for all known SNPs from UCSC or NCBI websites and then convert them using information from the SNP annotation files.