Email updates

Keep up to date with the latest news and content from BMC Medical Genetics and BioMed Central.

This article is part of the supplement: The Framingham Heart Study 100,000 single nucleotide polymorphisms resource

Open Access Research

Genome-wide association with diabetes-related traits in the Framingham Heart Study

James B Meigs1*, Alisa K Manning2, Caroline S Fox3, Jose C Florez4, Chunyu Liu2, L Adrienne Cupples2 and Josée Dupuis2

Author Affiliations

1 General Medicine Division, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA

2 Department of Biostatistics, Boston University School of Public Health, Boston, MA, USA

3 The Division of Endocrinology, Diabetes, and Hypertension, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, and the National Heart, Lung, and Blood Institute's Framingham Heart Study, Framingham, MA, USA

4 Diabetes Unit, Department of Medicine and Center for Human Genetic Research, Massachusetts General Hospital and Harvard Medical School, Boston, MA, and Program in Medical and Population Genetics, Broad Institute of Harvard and MIT, Cambridge, MA, USA

For all author emails, please log on.

BMC Medical Genetics 2007, 8(Suppl 1):S16  doi:10.1186/1471-2350-8-S1-S16

Published: 19 September 2007

Abstract

Background

Susceptibility to type 2 diabetes may be conferred by genetic variants having modest effects on risk. Genome-wide fixed marker arrays offer a novel approach to detect these variants.

Methods

We used the Affymetrix 100K SNP array in 1,087 Framingham Offspring Study family members to examine genetic associations with three diabetes-related quantitative glucose traits (fasting plasma glucose (FPG), hemoglobin A1c, 28-yr time-averaged FPG (tFPG)), three insulin traits (fasting insulin, HOMA-insulin resistance, and 0–120 min insulin sensitivity index); and with risk for diabetes. We used additive generalized estimating equations (GEE) and family-based association test (FBAT) models to test associations of SNP genotypes with sex-age-age2-adjusted residual trait values, and Cox survival models to test incident diabetes.

Results

We found 415 SNPs associated (at p < 0.001) with at least one of the six quantitative traits in GEE, 242 in FBAT (18 overlapped with GEE for 639 non-overlapping SNPs), and 128 associated with incident diabetes (31 overlapped with the 639) giving 736 non-overlapping SNPs. Of these 736 SNPs, 439 were within 60 kb of a known gene. Additionally, 53 SNPs (of which 42 had r2 < 0.80 with each other) had p < 0.01 for incident diabetes AND (all 3 glucose traits OR all 3 insulin traits, OR 2 glucose traits and 2 insulin traits); of these, 36 overlapped with the 736 other SNPs. Of 100K SNPs, one (rs7100927) was in moderate LD (r2 = 0.50) with TCF7L2 (rs7903146), and was associated with risk of diabetes (Cox p-value 0.007, additive hazard ratio for diabetes = 1.56) and with tFPG (GEE p-value 0.03). There were no common (MAF > 1%) 100K SNPs in LD (r2 > 0.05) with ABCC8 A1369S (rs757110), KCNJ11 E23K (rs5219), or SNPs in CAPN10 or HNFa. PPARG P12A (rs1801282) was not significantly associated with diabetes or related traits.

Conclusion

Framingham 100K SNP data is a resource for association tests of known and novel genes with diabetes and related traits posted at http://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?id=phs000007 webcite. Framingham 100K data replicate the TCF7L2 association with diabetes.