SUMMIT Diabetic Kidney Disease - Summary Level Data


We are releasing the summary data from our genome-wide meta-analyses of diabetic kidney disease (DKD) - which includes 7 binary phenotypes and, a continuous phenotype, eGFR - to enable other researchers to examine variants or loci of their interest in association with these traits.

DKD phenotypes were analysed separately in subjects with type 1 diabetes, in subjects with type 2 diabetes, and were also combined in an analysis of DKD in subjects with either T1D or T2D. There are 38 files in total that include summary statistics from 'all DKD', CKD, CKD+DKD, ESRD vs controls, ESRD vs non ESRD, 'late DKD', 'early DKD'/'microalbuminuria' and eGFR phenotypes. Each file includes summary statistics for over 5 million directly genotyped or imputed single nucleotide polymorphisms (SNPs) from the 1000G phase 1 March 2012 b37 imputation panel.

DKD phenotypes were assessed for association with each SNP using a logistic regression for binary phenotypes and a linear regression for eGFR against genotype using an additive genetic model corrected for age, sex and duration of diabetes. P values were derived from a linear mixed model that took relatedness into account as well as age, sex and duration of diabetes.

Dataset Details

The datasets were generated by performing an inverse-variance weighted meta-analysis of allelic effects and a sample size weighted meta-analysis of p values in up to 40,340 subjects with diabetes. Meta-analyses were performed separated in subjects with T1D and T2D, and were then combined yielding 38 meta-analyses in total. The phenotype definitions for each DKD phenotype analysed can be found in the following publications:

The files include the summary statistics from the combined analysis of discovery and replication studies. The replication studies for DKD in subjects with T2D included subjects of either European or Asian descent. Therefore, for certain phenotypes, the combined analysis in subjects of European descent are provided as well as the summary statistics estimated in subjects of either European or Asian descent. The replication cohorts for DKD in subjects with T2D were unable to define ESRD vs controls so the discovery summary statistics are provided.

Phenotype T1D T2D T1D/T2D
European Multi-Ethnic European Multi-Ethnic European Multi-Ethnic
ESRD vs controls
ESRD vs non ESRD
Late DKD
Early DKD/Microalbuminuria

Acknowledging The Data

When using data from the downloadable meta-analyses results please acknowledge the source of the data as follows:

Data on diabetic kidney disease has been contributed by the SUMMIT Consortium and has been downloaded from

In addition to the above acknowledgement, please cite the papers below:

The Genetic Landscape of Renal Complications in Type 1 Diabetes.

Sandholm et al, 2016

Journal of the American Society of Nephrology 28, 557-574.

PMID: 27647854

A Genome-Wide Association Study of Diabetic Kidney Disease in Subjects With Type 2 Diabetes.

van Zuydam et al, 2018

Diabetes 67, 1414-1427.

PMID: 29703844