Data Descriptor | Open | Published:

Sequence data and association statistics from 12,940 type 2 diabetes cases and controls

Scientific Data volume 4, Article number: 170179 (2017) | Download Citation

  • An Erratum to this article was published on 23 January 2018

Abstract

To investigate the genetic basis of type 2 diabetes (T2D) to high resolution, the GoT2D and T2D-GENES consortia catalogued variation from whole-genome sequencing of 2,657 European individuals and exome sequencing of 12,940 individuals of multiple ancestries. Over 27M SNPs, indels, and structural variants were identified, including 99% of low-frequency (minor allele frequency [MAF] 0.1–5%) non-coding variants in the whole-genome sequenced individuals and 99.7% of low-frequency coding variants in the whole-exome sequenced individuals. Each variant was tested for association with T2D in the sequenced individuals, and, to increase power, most were tested in larger numbers of individuals (>80% of low-frequency coding variants in ~82 K Europeans via the exome chip, and ~90% of low-frequency non-coding variants in ~44 K Europeans via genotype imputation). The variants, genotypes, and association statistics from these analyses provide the largest reference to date of human genetic information relevant to T2D, for use in activities such as T2D-focused genotype imputation, functional characterization of variants or genes, and other novel analyses to detect associations between sequence variation and T2D.

Metadata summary

Design Type(s)
  • individual genetic characteristics comparison design
  • parallel group design
  • data integration objective
Measurement Type(s)
  • genetic sequence variation analysis
Technology Type(s)
  • whole genome sequencing
  • exome sequencing
Factor Type(s)
  • ethnic group
Sample Characteristic(s)
  • Homo sapiens
  • Finland
  • Germany
  • United Kingdom
  • Sweden
  • United States of America
  • South Korea
  • Singapore
  • Israel

Download metadata file

Machine-accessible metadata file describing the reported data (ISA-tab format)

Background & Summary

Genome wide association studies (GWAS) have provided a valuable but incomplete window into the genetic basis of type 2 diabetes (T2D)1. Common (minor allele frequency [MAF]>5%) variants at over 100 loci have been robustly associated with disease risk, but most have not yet been translated to causal variants, effector transcripts, or disease mechanisms2. Because common variants from GWAS have modest effect sizes, and because those previously published explain in aggregate only 10–15% of the genetic basis of T2D3, it has been hypothesized that variants unexplored by GWAS might have a greater impact on efforts to understand or treat T2D4,5.

To produce a more complete catalogue of rare, low-frequency, and common variants, the GoT2D and T2D-GENES consortia analysed whole-exome and genome sequence data in up to 12,940 individuals (6,504 T2D cases and 6,436 controls; Fig. 1a, Tables 1 and 2)3. First, to interrogate lower-frequency (MAF>0.5%) variation genome-wide, 2,657 Northern and Central European individuals were selected by GoT2D (1,326 cases, 1,331 controls) and characterized with a combination of low-pass (~5x) whole-genome sequencing, deep (~82x) whole-exome sequencing, and high-density (2.5M) SNP genotyping. Genetic variants from these assays were incorporated into a phased integrated panel (WGS panel), capturing an estimated 99% of variants, genome-wide, present in more than 0.5% of individuals (Table 3). Second, additional individuals from 10 cohorts spanning five ethnicities (European, Hispanic, South Asian, East Asian, and African American) were characterized by deep (~82x) whole-exome sequencing by T2D-GENES. The resultant T2D-GENES exome sequence data were combined with the GoT2D exome sequence data to produce a second panel of variation (WES panel), capturing an estimated 99.7% of coding variants present in more than 0.5% of the combined 12,940 individuals (Tables 4 and 5).

Figure 1: Overview of data and analysis generation.
Figure 1

Shown is a flowchart for variant calling, quality control, and propagation of variants in both the WGS panel and WES panel. (a) Variant calling and quality control in the WGS and WES panels. Individuals were characterized with one or more sequencing and genotyping technologies, and then individuals and variants were excluded based on quality control metrics. The final WGS panel consists of data from 2,857 individuals and 28.5M variants, while the final WES panel consists of data from 13,008 individuals and 3.04M variants. (b) Assessing variants in larger sample sizes. Non-coding variants from the WGS panel were studied via statistical imputation in 44,414 additional individuals from cohorts within the DIAGRAM consortium. Coding variants from the WES panel were genotyped on the exome array in 79,854 additional individuals. Modified from Extended Data Fig. 1 of Fuchsberger et al.8.

Table 1: Summary of studies included in WGS panel.
Table 2: Summary of studies included in WES panel.
Table 3: Summary of variants in the WGS panel.
Table 4: Summary of variant annotations in the WES panel.
Table 5: Summary of coding variant frequencies in the WES panel.

Each variant was tested for association with T2D under an additive genetic model. To increase power, variants were then assessed in larger sample sizes via one of two means (Fig. 1b). Coding variants were analysed in 79,854 additional individuals (28,305 T2D cases, 51,549 controls) via the Illumina Exome Array, which captures 81.6% of European MAF>0.5% coding variants in the WES panel. Non-coding variants (and coding variants absent from the Exome Array) were analysed in up to 44,414 additional individuals (11,645 cases, 32,769 controls) via statistical genotype imputation; after quality control, this analysis included 89% of variants observed in three or more individuals in the WGS panel. Each variant was tested for association with T2D in the additional individuals, under an additive genetic model, and association statistics were then combined with those from the sequence data via meta-analysis.

Collectively, these analyses suggest a limited role for low-frequency variation in the genetic basis of T2D3. However, they also demonstrate an ability to identify novel hypotheses about the effects of gene inactivation6, a resource of coding variants for calibrating cellular assays7, and a catalogue of noncoding variants for use in statistical or functional fine mapping of GWAS signals3. The WGS panel also provides a novel resource for genotype imputation8, with increased resolution for T2D-specific variants relative to the 1000 Genomes (1000G) reference panel, as well as a means to calibrate simulation-based models of population history9 or disease genetic architecture10.

Methods

These methods are a modified version of the descriptions contained in Fuchsberger et al.3.

Ethics statement

All human research was approved by the relevant institutional review boards and conducted according to the Declaration of Helsinki. All participants provided written informed consent.

WGS GoT2D integrated panel generation

Ascertainment of individuals

Individuals were sampled from four studies: the Finland-United States Investigation of NIDDM Genetics (FUSION) Study (493 cases, 486 controls), KORA (101 cases, 104 controls), the UKT2D Consortium (322 cases, 322 controls), and the Malmö-Botnia Study (410 cases, 419 controls). All individuals were of Northern or Central European ancestry. Cases were preferentially lean, had (relatively) early onset T2D, or had a familial history of T2D; controls by comparison were preferentially overweight or had low fasting glucose levels11. To decrease the likelihood of selecting T2D cases who in fact had type 1 diabetes (T1D) or monogenic forms of diabetes (such as Maturity Onset Diabetes of the Young), cases with an age of diagnosis below 35, testing positive for GAD antibodies, or with a first-degree relative known to have T1D were not included. Statistics of the 2,657 individuals ultimately included in the association analysis are provided in Table 1.

Many of these individuals were measured for cardiometabolic phenotypes other than T2D, including glucose and insulin, anthropometrics, lipids, and blood pressure (Table 6).

Table 6: Additional cardiometabolic phenotypes measured in individuals included in the WGS and WES panels.

DNA sample preparation

De-identified DNA samples were sent to the Broad Institute in Cambridge, MA, USA (Malmö-Botnia and FUSION), the Wellcome Trust Centre for Human Genetics in Oxford, UK (UKT2D), or the Helmholtz Zentrum München in Germany (KORA). DNA quantity was measured by Picogreen (all samples) to ensure sufficient total DNA and minimum concentrations for downstream experiments. Samples (Malmö-Botnia, FUSION, UKT2D) were then genotyped on a Sequenom iPLEX assay for a set of 24 SNPs (one X chromosome and 23 autosomal SNPs), with only samples with high-quality genotypes advanced for subsequent sequencing or genotyping.

Exome sequencing

Genomic DNA was sheared, end repaired, ligated with barcoded Illumina sequencing adapters, amplified, size selected, and subjected to in-solution hybrid capture using the Agilent SureSelect Human All Exon v2.0 (Malmö-Botnia, FUSION, UK2T2D) or v3.0 (KORA) bait set (Agilent Technologies, USA). Resulting Illumina exome sequencing libraries were qPCR quantified, pooled, and sequenced with 76-bp paired-end reads using Illumina GAII or HiSeq 2,000 sequencers to ~82-fold mean coverage.

Genome sequencing

Whole-genome Illumina sequencing library construction was performed as described for exome sequencing above, except that genomic DNA was sheared to a larger target size and hybrid capture was not performed. The resulting libraries were size selected to contain fragment insert sizes of 380 bp±20% (Malmö-Botnia, FUSION, KORA) and 420 bp±25% (UKT2D) using gel electrophoresis or the SAGE Pippin Prep (Sage Science, USA). Libraries were qPCR quantified, pooled, and sequenced with 101-bp paired-end reads using Illumina GAII or HiSeq 2,000 sequencers to 5-fold mean coverage.

HumanOmni2.5 array genotyping and quality control (QC)

SNP array genotyping was performed by the Broad Genetic Analysis Platform. DNA samples were placed on 96-well plates and assayed using the Illumina HumanOmni2.5-4v1_B SNP array. Genotypes were then called using Illumina GenomeStudio v2010.3 with default clusters. SNPs with GenTrain score <0.6, cluster separation score <0.4, or call rate <97% were considered technical failures at the genotyping laboratory and excluded from further analysis. Next, 85 individuals with a genotype call rate below 98%, low genetic fingerprint (24-marker panel) concordance, or estimated gender discordance were excluded from further analysis. Finally, SNPs monomorphic across all individuals, failed by the 1000G Omni 2.5 QC filter, or with Hardy-Weinberg equilibrium P<10−6 were excluded from analysis.

Processing, quality control, and variant calling of sequence data

Sequence data were processed and aligned to the human genome (build hg19) using the Picard (http://broadinstitute.github.io/picard/), BWA12, and GATK13,14 software packages, following best-practice pipelines.

Sequencing coverage of each individual was computed based on the fraction of target bases with >20 reads aligned (exome sequencing) or average number of reads aligned across all bases genome-wide (genome sequencing). Based on these metrics, we excluded from further analysis exome sequence data (from 151 individuals) with coverage ≤20x in >20% of the target bases and genome sequence data (from 68 individuals) with average coverage ≤5x.

Possible DNA contamination of sequence data was assessed using verifyBamID15, either by direct comparison of sequence and HumanOmni2.5 array genotypes (where available) or by indirect estimates of contamination based on HumanOmni2.5 array allele frequencies. DNA samples with estimated contamination >2% using either method were excluded from further analysis (data from 7 individuals in the exome sequencing dataset and 59 individuals in the genome sequencing dataset). Uncontaminated DNA sample swaps were also detected via comparison of sequence and array data and corrected prior to variant calling.

To identify single nucleotide variants (SNVs) from the whole-genome sequence data, we used two independent SNV calling pipelines: GotCloud16 and the GATK UnifiedGenotyper14. We merged unfiltered SNV calls across the two call sets and then processed the merged site list through the SVM and VQSR filtering algorithms implemented by those pipelines. SNVs that failed both filtering algorithms were excluded from further analysis. To identify SNVs from the whole-exome sequence data, we used the GATK UnifiedGenotyper best-practices pipeline14.

To identify short insertions and deletions (indels) from the whole-genome sequence data, we called variants using the GATK UnifiedGenotyper best-practices pipeline. Because indels are known to have high false positive rates17, we applied more stringent criteria for indel QC than for SNV QC, excluding indels that failed either the SVM or VQSR filtering algorithms. To identify indels from the whole-exome sequence data, we used the GATK UnifiedGenotyper best-practices pipeline14.

To identify structural variants (SVs, or >100-bp deletions) from the whole-genome sequence data, we used GenomeSTRiP18. To increase sensitivity after initial discovery of SVs, we merged the discovered sites with deletions identified in 1,092 sequenced individuals from the 1000G Project17 and then genotyped the merged site lists across the whole-genome sequenced individuals. After applying the default filtering implemented in GenomeSTRiP, pass-filtered sites variable in any of the individuals were identified as candidate variant sites. Among these candidate sites, we excluded variants in known immunoglobin loci to reduce the impact of possible cell-line artifacts. We did not call SVs from the whole-exome sequence data.

Integrated panel generation

We merged variants discovered from the three experimental platforms into one site list. For individuals who had data from each of the three platforms, we then calculated genotype likelihoods across all sites separately by platform: for the whole-genome sequence data, we used GotCloud; for the exome sequence data, we used the GATK UnifiedGenotyper; and for the HumanOmni2.5 data, we converted hard genotype calls into genotype likelihoods assuming a genotype error rate of 10−6. If a site was not assayed by one of the three platforms, it was ignored in likelihood calculation for that platform.

We then calculated combined genotype likelihoods as the product of the genome, exome, and HumanOmni2.5 likelihoods, assuming independence across platforms. Following a strategy originally developed for the 1000G Phase 1 project17, we then phased the integrated likelihoods using Beagle19 (with 10,000 SNVs per chunk and 1,000 overlapping SNVs between consecutive chunks) and refined phased genotypes using Thunder20 as implemented in GotCloud (with 400 states).

Using the genotypes from the integrated panel, we performed principal component analysis (PCA) separately for each of the three variant types (SNVs, indels, SVs), using EPACTS on an LD-pruned (r2<0.20) set of MAF>0.01 autosomal variants (with variants in large high-LD regions21,22 or with Hardy-Weinberg P<10−6 removed). Inspecting the first ten PCs for each variant type, we identified 43 outlier individuals based on PCs from SNVs and indels and 136 additional outliers based on PCs from SVs; these 179 individuals were excluded from further analysis. Additionally, 38 individuals with close relationships with other study individuals (estimated genome-wide identity-by-descent proportion of alleles shared >0.20) were excluded from further analysis.

The final WGS panel contains genotypes from 2,874 individuals at 26.85M SNVs, 1.59M indels, and 11.88 K SVs. The final analysis set includes genotypes from 2,657 individuals at 26.20M SNVs, 1.50M indels, and 8.88K SVs SVs (Table 3).

WES (GoT2D+T2D-GENES Multiethnic) panel generation

Ascertainment of individuals

In addition to the individuals within the WGS panel, additional individuals, 10,242 of which were included in the final analysis, were chosen for whole-exome sequencing from 10 studies: the Jackson Heart Study (500 African-American cases, 526 matched controls), the Wake Forest School of Medicine Study (518 African-American cases, 530 matched controls), the Korea Association Research Project (526 East-Asian cases, 561 matched controls), the Singapore Diabetes Cohort Study and Singapore Prospective Study Program (486 East-Asian cases, 592 matched controls), Ashkenazi (506 European cases, 355 matched controls), the Metabolic Syndrome in Men (METSIM) Study (484 European cases, 498 matched controls), the San Antonio Mexican American Family Studies (272 Hispanic cases, 218 matched controls), the Starr County Texas study (749 Hispanic cases, 704 matched controls), the London Life Sciences Population (LOLIPOP) Study (531 South-Asian cases, 538 matched controls), and the Singapore Indian Eye Study (563 South-Asian cases, 585 matched controls). Potential T1D or MODY cases were excluded via similar approaches as for the whole-genome sequencing experiment. Statistics of the individuals ultimately included in the association analysis are provided in Table 2.

As for the WGS panel, many individuals were measured for additional cardiometabolic phenotypes (Table 6).

Exome sequencing

DNA samples were obtained and sequenced in the same manner as described for the exome sequencing component of the WGS panel.

Processing, QC, and variant calling

As for the exome sequence data within the WGS panel, sequence data for the WES panel were processed and aligned to the human genome (build hg19) using the Picard, BWA12, and GATK13,14 software packages and best-practice pipelines. Genotype likelihoods were computed controlling for contamination. Hard calls (the GATK-called genotypes but set as missing at a genotype quality [GQ] <20 threshold14) and dosages (the expected value of the genotype, defined as Pr(RX|data)+2Pr(XX|data), where R is the reference and X the alternative allele) were computed for each individual at each variant site. Hard calls were used only for quality control, while dosages were used in downstream association analyses. Multi-allelic SNVs and indels were dichotomized by collapsing alternate alleles into one category.

Individuals were excluded from analysis if they were outliers on one of multiple metrics: poor array genotype concordance (where available), high number of variant alleles or singletons, high or low allele balance (average proportion of non-reference alleles at heterozygous sites), or excess mean heterozygosity or ratio of heterozygous to homozygous genotypes. Within this reduced set of individuals, we then further excluded variants based on hard call rate (<90% in any cohort), deviation from Hardy-Weinberg equilibrium (P<10−6 in any ancestry group), or differential call rate between T2D cases and controls (P<10−4 in any ancestry group).

The final WES panel contains genotypes for 13,008 individuals at 2.93M SNVs and 111.9 K indels. The set ultimately included in coding variant association analysis (after removal of individuals with close relatives or of uncertain ancestry) contains genotypes for 12,940 individuals at 2.89M SNVs and 110.2 K indels (Tables 4 and 5).

Assaying variants in larger sample sizes

Imputation from the WGS panel

We carried out genotype imputation, using existing SNP array data, from the WGS panel into 44,414 individuals (11,645 T2D cases and 32,769 controls) from 13 studies participating in the DIAGRAM consortium. Each study performed quality control independently. A more detailed description of the analyzed individuals is available elsewhere3.

Exome array genotyping from the WES panel

We considered 28,305 T2D cases and 51,549 controls from 13 studies of European ancestry, each genotyped with the Illumina exome array. Studies independently called genotypes using the Illumina GenCall algorithm (http://www.illumina.com/Documents/products/technotes/technote_gencall_data_analysis_software.pdf ) or Birdseed23. Individuals were excluded if they had a low call rate (<99%), excess heterozygosity, high singleton counts, evidence of non-European ancestry, discrepancy between recorded and genotyped sex, or discordance with prior SNP array or genotyping platform fingerprint data (where available). Variants were excluded if they had a low call rate (<99%), deviation from Hardy-Weinberg equilibrium (P<10−6), GenTrain score <0.6, cluster separation score <0.4, or a suspect intensity plot based on manual inspection. After quality control, missing genotypes were re-called using zCall24, and additional quality control was performed to exclude poorly genotyped individuals (call rate <99% or excess heterozygosity) or variants (call rate <99%). A more detailed description of the analyzed individuals is available elsewhere3.

Association analysis

WGS panel single variant analysis

For each variant in the WGS panel, we tested for association between genotype and T2D in the 2,657 sequenced individuals. We used a logistic regression framework (assuming an additive genetic model) with the Firth bias-corrected likelihood ratio test25,26 to test for significance. Tests were adjusted for sex, the first two PCs computed based on genotypes from the HumanOmni2.5M array, and an indicator function for observed temporal stratification based on sequencing date and center.

Analysis of imputed datasets

In each of the thirteen studies within which variants from the WGS panel were imputed, SNVs with minor allele count (MAC)≥1 were tested for T2D association under an additive genetic model. Association tests were adjusted for study-specific covariates and performed using either the Firth, likelihood ratio, or score tests as implemented in EPACTS (https://genome.sph.umich.edu/wiki/EPACTS) or SNPTEST27. Residual population stratification for each study was accounted for using genomic control28. Association statistics were then combined across studies, using a fixed-effects sample-size weighted meta-analysis as implemented in METAL29.

WES panel single variant analysis

For each variant in the WES panel, we tested for association between genotype and T2D in the 12,940 sequenced individuals. We computed separate association statistics for each ancestry group using EMMAX30. Additionally, we performed association tests using the Wald statistic, adjusting for ethnic-specific principal components after exclusion of related individuals. For each test, we calculated genomic control inflation factors and corrected association summary statistics (P-values and standard errors) to account for residual population structure.

We subsequently performed a fixed-effects meta-analysis of ancestry-specific association summary statistics for each variant using (i) a sample-size weighting of P-values from the EMMAX analysis and (ii) an inverse-variance weighting of effect size estimates from the Wald analysis. For the final results, P-values were taken from the EMMAX analysis, and effect size estimates from the Wald analysis.

Analysis of exome array datasets

In each study within which exome array genotyping was applied, variants were tested for association with T2D via both the EMMAX and Wald tests. For the Wald test, related individuals were excluded and statistics were adjusted for study-specific principal components. For each study, P-values and standard errors were corrected based on the calculated genomic control inflation factor.

Variants were then combined via a fixed-effects meta-analysis. EMMAX P-values were combined via a sample-size weighted analysis, and Wald effect sizes were combined via an inverse-variant weighted analysis. For the final results, P-values were taken from the EMMAX analysis, and effect sizes were estimated from the Wald analysis.

Gene-level analysis

We first generated four variant lists (‘masks’) based on functional annotations and observed allele frequencies. Annotations were computed based on transcripts in ENSEMBL 66 (GRCh37.66) using CHAoS v0.6.3, SnpEFF v3.131, and VEP v2.732. We then identified variants predicted by at least one of the three algorithms in at least one mapped transcript to be protein-truncating ('for example, nonsense, frameshift, essential splice site), denoted PTVs, or other protein-altering (for example, missense, in-frame indel, non-essential splice site), denoted missense. We additionally used a previously described procedure33 to identify subsets of missense variants bioinformatically predicted to be deleterious: those annotated as damaging by each of Polyphen2-HumDiv, PolyPhen2-HumVar, LRT, Mutation Taster, and SIFT were considered to meet ‘strict’ criteria, while those annotated as damaging by one of these algorithms was considered to meet ‘broad’ criteria. We then calculated the MAF of each variant based on the highest frequency across each of the five ancestry groups. We finally combined these annotations to produce four masks: the PTV-only mask included PTVs, the PTV+NSstrict mask included variants in the PTV-only mask as well as those meeting ‘strict’ criteria for deleteriousness, the PTV+NSbroad mask included variants in the PTV-only mask as well as those with MAF<1% meeting ‘broad’ criteria for deleteriousness, and the PTV+missense mask included variants in the PTV+NSbroad mask as well as those with MAF<1% annotated as missense.

We performed gene-level analysis using the MetaSKAT software package (v0.32)34, employing the SKAT v0.93 library to perform a SKAT-O35 analysis within each ancestry group as well as across all ancestry groups via meta-analysis. Within each ancestry group, we assumed homogenous allele frequencies and genetic affects and adjusted for ethnic-specific axes of genetic variation after exclusion of 96 related individuals. For the meta-analysis, we used the MetaSKAT option to analyze genotype-level data, allowing for heterogeneity of allele frequencies and genetic effects between ancestry groups. All analyses were completed using the recommended ρ vector for SKAT-O: (0, 0.12, 0.22, 0.32, 0.52, 0.5, 1).

Code availability

All analyses were performed using publically available software packages, using versions and parameters as described above.

Data Records

Genotypes and phenotypes from the WGS and WES panels are available at the European Genome-phenome Archive (EGA, Data Citation 1: The European Genome-phenome Archive EGAS00001001459 and Data Citation 2: The European Genome-phenome Archive EGAS00001001460) and the database of Genotypes and Phenotypes (dbGAP, Data Citation 3: dbGAP phs001097.v1.p1 to Data Citation 12: dbGAP phs000840.v1.p1).

The data in EGA are covered under a single data use agreement, which complies with all of the cohort-specific data use restrictions. While this does limit data access according to the criteria of the most restrictive cohort, it is the only mechanism through which the entire WES and WGS panels are available to investigators. Additionally, the EGA contains data from one cohort that could not be released to a US-based repository. To download either the WGS or WES panel from the EGA, investigators must obtain approval from a data access committee (DAC, t2dgenes-got2d-dac@broadinstitute.org) to analyze data from all cohorts included in the study. The requester will receive an application packet that includes a project proposal document and a Data Transfer Agreement (DTA). The requester must then provide to the DAC a short description of their study, the proposed use of the data, an approval from the Institution's IRB, and a signed DTA. Assuming IRB approval and an executed DTA, the process for obtaining final approval from the DAC takes 4–6 weeks. Once approved, investigators can download either a single VCF file with genotypes from all individuals in the WGS panel (Data Citation 1: The European Genome-phenome Archive EGAS00001001459) or a single VCF file with genotypes from all individuals in the WES panel (Data Citation 2: The European Genome-phenome Archive EGAS00001001460).

If investigators cannot obtain approval to analyze all cohorts in the WES panel (e.g., commercial uses) they can download cohort-specific data from dbGAP. Each cohort in dbGAP is subject to distinct data use restrictions, and investigators can obtain separate VCF files, as well as the raw sequence reads, for each of the cohorts.

The WGS and WES panels are accompanied by exclusion lists of variants and individuals (Data Citation 1: The European Genome-phenome Archive EGAS00001001459 and Data Citation 2: The European Genome-phenome Archive EGAS00001001460). The WGS VCF file contains data from the full set of 2,874 individuals and 28.45M variants that passed QC, with additional lists provided containing the 2,657 individuals and 17.69M variants included in association analysis. The WES VCF file contains the full set of 13,008 individuals that passed QC, with additional lists provided containing the 3.04M variants that passed QC (the VCF includes a small number of variants that failed QC) and the 12,940 samples and 2.93M variants included in association analysis. Additionally, the WES dataset includes lists of variants included in gene-level analysis for each of the four analyzed masks.

Five datasets of association statistics are also available for download. Association statistics for variants in the WGS panel are available from the whole-genome sequenced individuals or from those with imputed genotypes (Data Citation 1: The European Genome-phenome Archive EGAS00001001459). Association statistics in the WES panel are available from the whole-exome sequenced individuals or from those genotyped on the exome array (Data Citation 2: The European Genome-phenome Archive EGAS00001001460). Additionally, gene-level association statistics from the whole-exome sequenced individuals are available for each of the four variant masks (Data Citation 2: The European Genome-phenome Archive EGAS00001001460).

A description of the datasets is available in Table 7. All association statistics are also available for browsing and searching via the public Type 2 Diabetes Knowledge Portal at www.type2diabetesgenetics.org. Through the portal, users can construct queries to find variants satisfying specified annotations and association thresholds, both across the WES and WGS analyses as well as other GWAS datasets. Users can also dynamically construct a set of variants within a gene and obtain a P-value from aggregate association analysis within the WES individuals.

Table 7: Summary of datasets.

Technical Validation

Evaluation of variants in the WGS panel

We evaluated the variant sensitivity (fraction of true variant sites detected) of the WGS panel, based on the 2,538 individuals with data from all three experimental platforms (low-pass whole-genome sequencing, whole-exome sequencing, and HumanOmni2.5M array genotyping). To assess the sensitivity of low-pass whole-genome sequencing alone, we computed the fraction of variants detected from whole-exome sequencing that were also detected by low-pass whole-genome sequencing. Sensitivity estimates were 99.8, 99.0, and 48.2% for common (MAF>5%), low-frequency (0.5%<MAF<5%), and rare (MAF<0.5%) SNVs, respectively, and >99.9, 93.8, and 17.9% for common, low-frequency, and rare short indels, respectively. We also assessed the coding SNV sensitivity of low-pass whole-genome sequence data combined with exome sequence data, based on the proportion of HumanOmni2.5 SNVs detected by either sequencing platform. Because HumanOmni2.5 SNVs are enriched for common variants, we calculated an averaged sensitivity at each allele count, weighted by the number of exome-detected variants given the allele count. Sensitivity estimates were 99.9, 99.7, and 83.9% for common, low-frequency, and rare variants, respectively. These sensitivity estimates provide lower bounds on the sensitivity of the full WGS panel, which combines HumanOmni2.5 SNP array data as well as the two types of sequence data (Fig. 2a).

Figure 2: Summary of key quality control metrics for WGS and WES panels.
Figure 2

We computed several metrics to verify the sequencing accuracy of the study individuals. (a) Estimates of sensitivity of WGS panel. Shown is the fraction of variants, as a function of minor allele count in the WGS sequenced individuals, estimated as included in the WGS panel. Green circles show the total fraction of variants; green crosses show the fraction of variants for hypothetical variants with a T2D odds ratio of 5 (because T2D cases are overrepresented in our sample, the sensitivity to detect risk variants is increased). For comparison, shown are the fraction of variants that are included in the 1000G Phase 1 dataset (blue circles) or HapMap panel (red circles). (b) Distribution of minor alleles carried by individuals in the WES panel. For different populations within the WES panel, the distribution of minor alleles carried is plotted across all individuals. A normal distribution indicates a lack of systematic sequencing artefacts for any one individual, at least according to this metric. Afr-Am: African American. (c) Comparison of principal components computed from SNPs and indels versus indels alone. We calculated principal components for the European individuals in the WES panel using all variants in the panel and then again using only indels. Adapted from Supplementary Tables 5 and 6, and Fig. 1a, in Fuchsberger et al.8.

We further evaluated the genotype accuracy of the WGS panel for each of the three classes of variant (SNVs, indels, and SVs). Across chromosome 20, concordance of low-pass whole-genome-sequence-based SNV genotypes with exome-sequence-based genotypes was 99.86%, with homozygous reference, heterozygous, and homozygous non-reference concordances of 99.97, 98.34, and 99.72%, respectively. Concordance between exome-sequence-based SNV genotypes and HumanOmni2.5 genotypes was 99.4%, with homozygous reference, heterozygous, and homozygous non-reference concordances of 99.97, 99.69, and 99.88%, respectively. For indels genotyped with both low-pass whole-genome-sequence data and exome-sequence data, concordance was 99.4%, with homozygous reference, heterozygous, and homozygous non-reference concordances of 99.8, 95.8, and 98.6%, respectively.

To evaluate the genotype accuracy of SVs detected from the low-pass whole-genome sequence data, we took advantage of the 181 individuals in our study who were previously included in the WTCCC array-CGH based structural variant detection experiment36. Taking the WTCCC data as a gold standard, we estimated genotype accuracy across 1,047 overlapping SVs (with reciprocal overlap >0.8) genome-wide. The overall genotype concordance was 99.8%, with homozygous reference, heterozygous, and homozygous non-reference concordances of 99.9, 99.6, and 99.7%, respectively.

Evaluation of variants in the WES panel

We assessed the overall sequencing quality of individuals in the WES panel by computing distributions of global statistics, stratified by reported ancestry (Fig. 2b). After quality control, the number of non-reference variants, mean heterozygosity, and average allele balance (fraction of non-reference reads at heterozygous sites) per individual approximately matched a Gaussian distribution within each ancestry. Concordance between genotypes from exome-sequence data and those from independent SNP arrays was above 99% for the vast majority of individuals, with non-reference concordance above 99.5% for individuals genotyped on the (highest-quality) OMNI array.

We also assessed bulk properties of indels within the WES panel. The length distribution of indels showed an excess of variants with lengths a multiple of three, as expected. Additionally, principal components computed from indels alone closely matched those computed from SNVs and indels together (Fig. 2c).

Evaluation of imputation from the WGS panel

We computed the mean imputation quality as measured by the average squared correlation between imputed genotypes and actual genotypes from leave-one-out cross-validation analysis. For variants of allele count ≈100 or above in the WGS panel (corresponding to a frequency of 1.8%), average r2 values were in excess of 0.8 for Finnish individuals and in excess of 0.6 for British individuals (Fig. 3a).

Figure 3: Completeness of additional variant genotyping.
Figure 3

We calculated the fraction of variants in the WGS and WES panel that were captured via either imputation or exome array genotyping, respectively. (a) The mean imputation quality of variants in the WGS panel, as a function of their allele count in the WGS panel. Green circles show imputation quality in Finnish individuals, while green crosses show imputation quality in British individuals. For comparison, blue circles and crosses show imputation quality using the 1000G Phase 1 dataset as a reference panel (instead of the WGS panel). (b) The number of coding variants in the WES panel present on the exome array. Variants are stratified by annotation and frequency, and sensitivity calculations are shown for variants in each ancestry group as well as overall. Panel (b) is reproduced from Supplementary Fig. 17 in Fuchsberger et al.8.

Evaluation of exome array sensitivity

We assessed the overlap of variants present on the exome array with those observed in the WES panel. As the exome array primarily contains SNVs that are predicted to be protein altering, we focused on nonsense, essential splice site, and missense variants; only variants passing QC in both sequence and array data were included in the assessment. The fraction of variants in the WES panel on the exome array was highest for Europeans, at 81.6%, and lowest in African-Americans, at 49.0% (Fig. 3b).

Evaluation of association tests

We used the genetic power calculator (http://zzz.bwh.harvard.edu/gpc/) to estimate power to detect T2D association for each of the single variant analyses. All calculations assumed a T2D prevalence of 8%. Figure 4 shows power estimates under optimistic scenarios, in which a variant is present in the WGS panel (Fig. 4a), observed at equal frequency in all ancestries in the WES panel (Fig. 4b), imputed with high (r2=0.8) accuracy (Fig. 4c), or included on the exome array (Fig. 4d). We also computed power for less optimistic scenarios3; if, for example, a MAF 1% variant is present in only one ancestry, it must have an odds ratio of ~3.5 to achieve significance of P=10−4, rather than an odds ratio of ~1.8 were it to be present in all ancestries.

Figure 4: Power of single variant analysis in the WGS panel, WES panel, imputation, and exome array analyses.
Figure 4

Shown is the power to detect an association with variants of varying population frequencies and T2D odds ratios, at a relatively lenient significance level of α=10−4. Such a significance level would be insufficient to establish an association due to the burden of multiple testing, but lack of association at this significance level can place bounds on the maximum effect a variant has in the population. (a) Power for a variant of constant frequency and effect across all populations in the WGS panel. (b) Power for a variant of constant frequency and effect across all populations in the WES panel. (c) Power for a variant imputed from the WGS panel with imputation accuracy r2=0.8. (d) Power for a variant in both the WES panel and on the exome array. NA: number of affecteds (cases); NU: number of unaffecteds (controls); K: presumed prevalence of T2D in the population.

For gene-based tests in the WES panel, we used a simulated haplotype data set (http://cran.r-project.org/web/packages/SKAT/vignettes/SKAT.pdf) and estimated power as a function of (i) the phenotypic variance, under a liability scale, explained by additive genetic effects and (ii) the percentage of variants that were causal (50% or 100%). As for single-variant power calculations, we considered variants of constant frequency across all five ancestry groups, as well as variants specific to one ancestry group. These calculations suggested3 that, even under optimistic scenarios, genes must explain >1% of genetic variance in order to achieve a moderately significant (P<10−3) association in the WES panel.

To ensure that association statistics were well calibrated, we computed quantile-quantile (QQ) plots comparing observed statistics to those expected under the null distribution. The vast majority of statistics matched the expected distribution (suggesting good calibration of the association tests) with a deviation from the null for common variant associations from the WGS panel, imputed genotypes, and exome array genotypes (suggesting power to detect known positive control T2D-associated non-coding common variants).

Usage Notes

The WGS and WES panels may be useful for simulation-based approaches that require individual-level genotypes and phenotypes. In this case the full list of ‘QC+’ variants (not merely those included in the T2D analysis) should be used, as the association analysis omitted very rare variants that might be useful in other settings. The WGS panel can also serve as a reference panel for genotype imputation, particularly in cases where an excess of haplotypes from T2D cases are required. Although more recent and larger efforts such as the Haplotype Reference Consortium8 will provide greater imputation power for most use cases, the WGS panel is not restricted based on minor allele count and includes indels and SVs.

The most valuable data from the T2D-GENES and GoT2D studies are likely the catalogues of T2D association statistics for low-frequency and common variation. These statistics may prove useful for fine mapping or functional studies of T2D GWAS signals, in which enumeration of potential causal variants is required, or for ‘reverse genetic’ approaches, in which estimates of the phenotypic effects of variants with strong molecular effects are desired. For this usage, we advise investigators to query the T2D Knowledge Portal at www.type2diabetesgenetics.org as a first step, as its goal is to provide a simple and continuously updated means to query these and other association statistics. The portal is designed specifically for queries about individual variants or those within a single gene or genomic locus, as well as variant- or gene-level analyses for which investigators desire to adjust included variants, covariates, or individuals. Users should note that data from the T2D-GENES and GoT2D studies are included in, and thus should not be combined with data from, the Exome Aggregation Consortium (exac.broadinstitute.org).

Should investigators desire access to all association statistics, genome-wide, the files at EGA should be used (as the T2D Knowledge Portal does not support bulk download of association statistics). For single variant associations, a statistic in the largest available sample size should be used. Coding variant association statistics present in both the WES panel analysis and the exome chip analysis can be safely combined via meta-analysis, as can non-coding variant association statistics present in both the WGS panel analysis and the imputation-based analysis; statistics should not, however, be combined across the non-coding and coding variant analyses. Association results from the sequence or exome chip data should not need to be filtered, but it is advisable to filter results from the imputation-based analysis according to a threshold on imputation quality (e.g., r2>0.3). For gene-level analyses, investigators should first use the aggregate association statistics and then dissect results by examining variant-level statistics for each variant in the mask.

The pre-computed statistics should be sufficient for most investigators. Cases where recalculation of associations may be appropriate include (a) conditional analyses, such as variant association controlling for additional individual phenotypes or genotypes at different variants, (b) association analyses with phenotypes other than T2D, and (c) novel statistical tests. In these cases, usage of the tests and inclusion of the covariates described in the methods section is recommended, and only variants and individuals present in the final ‘QC+’ analysis lists should be included.

Additional information

How to cite this article: Flannick, J. et al. Sequence data and association statistics from 12,940 type 2 diabetes cases and controls. Sci. Data 4:170179 doi: 10.1038/sdata.2017.179 (2017).

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    et al. Finding the missing heritability of complex diseases. Nature 461, 747–753 (2009).

  2. 2.

    & Type 2 diabetes: genetic data sharing to advance complex disease research. Nature Reviews Genetics 17, 535–549 (2016).

  3. 3.

    et al. The genetic architecture of type 2 diabetes. Nature 536, 41–47 (2016).

  4. 4.

    & Common and rare variants in multifactorial susceptibility to common diseases. Nature Genetics 40, 695–701 (2008).

  5. 5.

    & Uncovering the roles of rare variants in common disease through whole-genome sequencing. Nature Reviews Genetics 11, 415–425 (2010).

  6. 6.

    et al. Loss-of-function mutations in SLC30A8 protect against type 2 diabetes. Nature Genetics 46, 357–363 (2014).

  7. 7.

    et al. Rare variants in PPARG with decreased activity in adipocyte differentiation are associated with increased risk of type 2 diabetes. Proceedings of the National Academy of Sciences of the United States of America 111, 13127–13132 (2014).

  8. 8.

    et al. A reference panel of 64,976 haplotypes for genotype imputation. Nature Genetics 48, 1279–1283 (2016).

  9. 9.

    et al. Simulation of Finnish population history, guided by empirical genetic data, to assess power of rare-variant tests in Finland. American Journal of Human Genetics 94, 710–720 (2014).

  10. 10.

    , , , & Evaluating empirical bounds on complex disease genetic architecture. Nature Genetics 45, 1418–1427 (2013).

  11. 11.

    et al. Power in the phenotypic extremes: a simulation study of power in discovery and replication of rare variants. Genetic Epidemiology 35, 236–246 (2011).

  12. 12.

    & Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760 (2009).

  13. 13.

    et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Research 20, 1297–1303 (2010).

  14. 14.

    et al. A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nature Genetics 43, 491–498 (2011).

  15. 15.

    et al. Detecting and estimating contamination of human DNA samples in sequencing and array-based genotype data. American Journal of Human Genetics 91, 839–848 (2012).

  16. 16.

    , , & An efficient and scalable analysis framework for variant extraction and refinement from population-scale DNA sequence data. Genome Research 25, 918–925 (2015).

  17. 17.

    et al. An integrated map of genetic variation from 1,092 human genomes. Nature 491, 56–65 (2012).

  18. 18.

    , , & Discovery and genotyping of genome structural polymorphism by sequencing on a population scale. Nature Genetics 43, 269–276 (2011).

  19. 19.

    & Rapid and accurate haplotype phasing and missing-data inference for whole-genome association studies by use of localized haplotype clustering. American Journal of Human Genetics 81, 1084–1097 (2007).

  20. 20.

    , , , & Low-coverage sequencing: implications for design of complex trait association studies. Genome Research 21, 940–951 (2011).

  21. 21.

    et al. Long-range LD can confound genome scans in admixed populations. American Journal of Human Genetics 83, 132–135, author reply 135-139 (2008).

  22. 22.

    Quality control for genome-wide association studies. Methods in Molecular Biology 628, 341–372 (2010).

  23. 23.

    et al. Integrated genotype calling and association analysis of SNPs, common copy number polymorphisms and rare CNVs. Nature Genetics 40, 1253–1260 (2008).

  24. 24.

    et al. zCall: a rare variant caller for array-based genotyping: genetics and population analysis. Bioinformatics 28, 2543–2545 (2012).

  25. 25.

    Bias reduction of maximum-likelihood-estimates. Biometrika 80, 27–38 (1993).

  26. 26.

    , , , & i. Recommended joint and meta-analysis strategies for case-control association testing of single low-count variants. Genetic Epidemiology 37, 539–550 (2013).

  27. 27.

    , , , & A new multipoint method for genome-wide association studies by imputation of genotypes. Nature Genetics 39, 906–913 (2007).

  28. 28.

    & Genomic control for association studies. Biometrics 55, 997–1004 (1999).

  29. 29.

    , & METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics 26, 2190–2191 (2010).

  30. 30.

    et al. Variance component model to account for sample structure in genome-wide association studies. Nature Genetics 42, 348–354 (2010).

  31. 31.

    et al. A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly 6, 80–92 (2012).

  32. 32.

    et al. The Ensembl Variant Effect Predictor. Genome Biology 17, 122 (2016).

  33. 33.

    et al. A polygenic burden of rare disruptive mutations in schizophrenia. Nature 506, 185–190 (2014).

  34. 34.

    , , & General framework for meta-analysis of rare variants in sequencing association studies. American Journal of Human Genetics 93, 42–53 (2013).

  35. 35.

    , & Optimal tests for rare variant effects in sequencing association studies. Biostatistics 13, 762–775 (2012).

  36. 36.

    Wellcome Trust Case Control Consortium. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 447, 661–678 (2007).

Download references

Data Citations

  1. 1.

    The European Genome-phenome Archive EGAS00001001459 (2016)

  2. 2.

    The European Genome-phenome Archive EGAS00001001460 (2016)

  3. 3.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs001097.v1.p1 (2016)

  4. 4.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs001099.v1.p1 (2016)

  5. 5.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs001098.v1.p1 (2016)

  6. 6.

    Duggirala, R. dbGAP phs000849.v1.p1 (2016)

  7. 7.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs001096.v1.p1 (2016)

  8. 8.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs001095.v1.p1 (2016)

  9. 9.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs001093.v1.p1 (2016)

  10. 10.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs001100.v1.p1 (2016)

  11. 11.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs001102.v1.p1 (2016)

  12. 12.

    Altshuler, D., Boehnke, M., McCarthy, M., & Florez, J. dbGAP phs000840.v1.p1 (2016)

Acknowledgements

Grant support and acknowledgments are listed in the Supplementary Information.

Author information

Author notes

    • Jason Flannick
    • , Christian Fuchsberger
    •  & Anubha Mahajan

    These authors contributed equally to this work.

    • David Altshuler
    • , Michael Boehnke
    •  & Mark I. McCarthy

    These authors jointly supervised this work.

    • Peter S. Chines
    •  & Hanna E. Abboud

    Deceased.

Affiliations

  1. Department of Molecular Biology, Massachusetts General Hospital, Boston, Massachusetts, USA

    • Jason Flannick
    • , Hyun Min Kang
    •  & David Altshuler
  2. Program in Medical and Population Genetics, Broad Institute, Cambridge, Massachusetts, USA

    • Jason Flannick
    • , Vineeta Agarwala
    • , Lizz Caulkins
    • , Ryan Koesterer
    • , Cecilia M. Lindgren
    • , Christopher Hartl
    • , Todd Green
    • , Alisa Manning
    • , Jason Carey
    • , George Grant
    • , Benjamin M. Neale
    • , Shaun Purcell
    • , Tõnu Esko
    • , Mauricio O. Carneiro
    • , Jared Maguire
    • , Ryan Poplin
    • , Khalid Shakir
    • , Timothy Fennell
    • , Mark DePristo
    • , Jacquelyn Murphy
    • , Robert Onofrio
    • , Eric Banks
    • , Stacey Gabriel
    • , David Altshuler
    • , Noël P. Burtt
    •  & Jose C. Florez
  3. Department of Biostatistics and Center for Statistical Genetics, University of Michigan, Ann Arbor, Michigan, USA

    • Christian Fuchsberger
    • , Tanya M. Teslovich
    • , Clement Ma
    • , Xueling Sim
    • , Thomas W. Blackwell
    • , Adam E. Locke
    • , Anne U. Jackson
    • , Jeroen R. Huyghe
    • , Heather M. Stringham
    • , Keng-Han Lin
    • , Ryan P. Welch
    • , Phoenix Kwan
    • , Goo Jun
    • , Goncalo Abecasis
    • , Laura J. Scott
    •  & Michael Boehnke
  4. Wellcome Trust Centre for Human Genetics, Nuffield Department of Medicine, University of Oxford, Oxford, UK

    • Anubha Mahajan
    • , Kyle J. Gaulton
    • , Loukas Moutsianas
    • , Davis J. McCarthy
    • , Manuel A. Rivas
    • , John R. B. Perry
    • , Neil R. Robertson
    • , N William Rayner
    • , Juan Fernandez Tajes
    • , Cecilia M. Lindgren
    • , Martijn van de Bunt
    • , Richard D. Pearson
    • , Ashish Kumar
    • , Yuhui Chen
    • , Teresa Ferreira
    • , Momoko Horikoshi
    • , Erik Ingelsson
    • , Inga Prokopenko
    • , Anna L. Gloyn
    • , Peter Donnelly
    • , Gilean McVean
    • , Andrew P. Morris
    •  & Mark I. McCarthy
  5. Harvard-MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA

    • Vineeta Agarwala
  6. Department of Statistics, University of Oxford, Oxford, UK

    • Davis J. McCarthy
    •  & Peter Donnelly
  7. Genetics of Complex Traits, University of Exeter Medical School, University of Exeter, Exeter, UK

    • John R. B. Perry
    • , Dorota Pasko
    • , Andrew R. Wood
    •  & Timothy M. Frayling
  8. MRC Epidemiology Unit, Institute of Metabolic Science, University of Cambridge, Cambridge, UK

    • John R. B. Perry
    • , Robert A. Scott
    • , Claudia Langenberg
    •  & Nicholas J. Wareham
  9. Department of Twin Research and Genetic Epidemiology, King's College London, London, UK

    • John R. B. Perry
    • , Massimo Mangino
    • , Gabriela L. Surdulescu
    • , Dylan Hodgkiss
    • , Kerrin S. Small
    •  & Timothy D. Spector
  10. Oxford Centre for Diabetes, Endocrinology and Metabolism, Radcliffe Department of Medicine, University of Oxford, Oxford, UK

    • Neil R. Robertson
    • , N William Rayner
    • , Martijn van de Bunt
    • , Nicola L. Beer
    • , Momoko Horikoshi
    • , Jonathan C. Levy
    • , Christopher J. Groves
    • , Matt Neville
    • , Fredrik Karpe
    • , Inga Prokopenko
    • , Katharine R. Owen
    • , Anna L. Gloyn
    •  & Mark I. McCarthy
  11. Department of Human Genetics, Wellcome Trust Sanger Institute, Hinxton, Cambridgeshire, UK

    • N William Rayner
    • , Aaron G. Day-Williams
    • , John Danesh
    • , Kathleen Stirrups
    • , Panos Deloukas
    • , Manjinder Sandhu
    • , Inês Barroso
    •  & Eleftheria Zeggini
  12. School of Computer Science, McGill University, Montreal, Quebec, Canada

    • Pablo Cingolani
  13. McGill University and Génome Québec Innovation Centre, Montreal, Quebec, Canada

    • Pablo Cingolani
    • , Yoshihiko Nagai
    •  & Rob Sladek
  14. Human Genetics Center, The University of Texas Graduate School of Biomedical Sciences at Houston, The University of Texas Health Science Center at Houston, Houston, Texas, USA

    • Heather M. Highland
  15. Department of Biostatistics, Boston University School of Public Health, Boston, Massachusetts, USA

    • Josee Dupuis
    • , Han Chen
    •  & Dennis Rybin
  16. National Heart, Lung, and Blood Institute's Framingham Heart Study, Framingham, Massachusetts, USA

    • Josee Dupuis
  17. Medical Genomics and Metabolic Genetics Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland, USA

    • Peter S. Chines
    • , Lori L. Bonnycastle
    • , Narisu Narisu
    • , Amy Swift
    •  & Francis S. Collins
  18. Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts, USA

    • Han Chen
    •  & Liming Liang
  19. Chronic Disease Epidemiology, Swiss Tropical and Public Health Institute, University of Basel, Basel, Switzerland

    • Ashish Kumar
  20. Institute of Genetic Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany

    • Martina Müller-Nurasyid
    • , Janina S. Ried
    • , Christian Gieger
    •  & Konstantin Strauch
  21. Department of Medicine I, University Hospital Grosshadern, Ludwig-Maximilians-Universität, Munich, Germany

    • Martina Müller-Nurasyid
  22. Chair of Genetic Epidemiology, IBE, Faculty of Medicine, LMU Munich, Germany

    • Martina Müller-Nurasyid
    •  & Konstantin Strauch
  23. DZHK (German Centre for Cardiovascular Research), partner site Munich Heart Alliance, Munich, Germany

    • Martina Müller-Nurasyid
    •  & Annette Peters
  24. The Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

    • Niels Grarup
    • , Jette Bork-Jensen
    • , Mette Hollensted
    • , Johanne Marie Justesen
    • , Anette P. Gjesing
    • , Torben Hansen
    •  & Oluf Pedersen
  25. Department of Medicine, Section of Genetic Medicine, The University of Chicago, Chicago, Illinois, USA

    • Eric R. Gamazon
    • , Hae Kyung Im
    •  & Nancy J. Cox
  26. Department of Statistics, Seoul National University, Seoul, Republic of Korea

    • Jaehoon Lee
    • , Iksoo Huh
    • , Yongkang Kim
    • , Selyeong Lee
    •  & Taesung Park
  27. Human Genetics Center, School of Public Health, The University of Texas Health Science Center at Houston, Houston, Texas, USA

    • Jennifer E. Below
    • , Taylor J. Maxwell
    • , Goo Jun
    •  & Craig L. Hanis
  28. Saw Swee Hock School of Public Health, National University of Singapore, National University Health System, Singapore, Singapore

    • Peng Chen
    • , Xu Wang
    • , Ching-Yu Cheng
    • , Chiea-Chuen Khor
    • , Wei Yen Lim
    • , Jianjun Liu
    • , Kee Seng Chia
    • , Yik Ying Teo
    •  & E. Shyong Tai
  29. Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts, USA

    • Jinyan Huang
    • , Frank B. Hu
    •  & Liming Liang
  30. Center for Genome Science, Korea National Institute of Health, Chungcheongbuk-do, Republic of Korea

    • Min Jin Go
    • , Bong-Jo Kim
    • , Young Jin Kim
    • , Juyoung Lee
    • , Bok-Ghee Han
    •  & Jong-Young Lee
  31. The Jackson Laboratory for Genomic Medicine, Farmington, Connecticut, USA

    • Michael L. Stitzel
  32. Departments of Computational Medicine & Bioinformatics and Human Genetics, University of Michigan, Ann Arbor, Michigan, USA

    • Stephen C. J. Parker
  33. Department of Clinical Sciences, Lund University Diabetes Centre, Genetic and Molecular Epidemiology Unit, Lund University, Malmö, Sweden

    • Tibor V. Varga
    •  & Paul W. Franks
  34. Department of Epidemiology, Colorado School of Public Health, University of Colorado, Aurora, Colorado, USA

    • Tasha Fingerlin
  35. Department of Endocrinology and Metabolism, Shanghai Diabetes Institute, Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Shanghai, China

    • Cheng Hu
    •  & Weiping Jia
  36. Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Singapore

    • Mohammad Kamran Ikram
    • , Tin Aung
    • , Ching-Yu Cheng
    • , Chiea-Chuen Khor
    •  & Tien Yin Wong
  37. Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, National University Health System, Singapore, Singapore

    • Mohammad Kamran Ikram
    • , Tin Aung
    • , Ching-Yu Cheng
    • , Chiea-Chuen Khor
    •  & Tien Yin Wong
  38. The Eye Academic Clinical Programme, Duke-NUS Graduate Medical School, Singapore, Singapore

    • Mohammad Kamran Ikram
    • , Tin Aung
    • , Ching-Yu Cheng
    •  & Tien Yin Wong
  39. Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, Republic of Korea

    • Min-Seok Kwon
    • , Joon Yoon
    •  & Taesung Park
  40. Department of Human Genetics, McGill University, Montreal, Quebec, Canada

    • Yoshihiko Nagai
    •  & Rob Sladek
  41. Research Institute of the McGill University Health Centre, Montreal, Quebec, Canada

    • Yoshihiko Nagai
  42. Department of Epidemiology and Biostatistics, Imperial College London, London, UK

    • Weihua Zhang
    • , Uzma Afzal
    • , Benjamin Lehne
    • , Marie Loh
    • , William R. Scott
    • , Paul Elliott
    •  & John C. Chambers
  43. Department of Cardiology, Ealing Hospital NHS Trust, Southall, Middlesex, UK

    • Weihua Zhang
    • , Sian-Tsung Tan
    • , Jaspal Singh Kooner
    •  & John C. Chambers
  44. Departments of Medicine and Genetics, Albert Einstein College of Medicine, New York, USA

    • Nir Barzilai
    •  & Gil Atzmon
  45. Department of Systems Pharmacology and Translational Therapeutics, University of Pennsylvania—Perelman School of Medicine, Philadelphia, Pennsylvania, USA

    • Benjamin F. Voight
  46. Department of Genetics, University of Pennsylvania—Perelman School of Medicine, Philadelphia, Pennsylvania, USA

    • Benjamin F. Voight
  47. Department of Medicine, University of Texas Health Science Center, San Antonio, Texas, USA

    • Christopher P. Jenkinson
    • , Hanna E. Abboud
    • , Sharon P. Fowler
    • , Farook Thameem
    • , Donna M. Lehman
    •  & Ralph A. DeFronzo
  48. Research, South Texas Veterans Health Care System, San Antonio, Texas, USA

    • Christopher P. Jenkinson
  49. Faculty of Health Sciences, Institute of Clinical Medicine, Internal Medicine, University of Eastern Finland, Kuopio, Finland

    • Teemu Kuulasmaa
    • , Johanna Kuusisto
    • , Alena Stančáková
    •  & Markku Laakso
  50. Kuopio University Hospital, Kuopio, Finland

    • Johanna Kuusisto
    •  & Markku Laakso
  51. Center for Genomics and Personalized Medicine Research, Wake Forest School of Medicine, Winston-Salem, North Carolina, USA

    • Maggie C. Y. Ng
    • , Nicholette D. Palmer
    • , Pamela J. Hicks
    •  & Donald W. Bowden
  52. Center for Diabetes Research, Wake Forest School of Medicine, Winston-Salem, North Carolina, USA

    • Maggie C. Y. Ng
    • , Nicholette D. Palmer
    • , Pamela J. Hicks
    •  & Donald W. Bowden
  53. Department of Biochemistry, Wake Forest School of Medicine, Winston-Salem, North Carolina, USA

    • Nicholette D. Palmer
    • , Pamela J. Hicks
    •  & Donald W. Bowden
  54. Centre for Research in Epidemiology and Population Health, Inserm U1018, Villejuif, France

    • Beverley Balkau
  55. German Institute of Human Nutrition Potsdam-Rehbruecke, Nuthetal, Germany

    • Heiner Boeing
  56. Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, Sweden

    • Vilmantas Giedraitis
    •  & Lars Lannfelt
  57. Centre for Chronic Disease Control, New Delhi, India

    • Dorairaj Prabhakaran
    •  & Shah B. Ebrahim
  58. The Charles Bronfman Institute for Personalized Medicine, The Icahn School of Medicine at Mount Sinai, New York, USA

    • Omri Gottesman
    • , Yingchang Lu
    • , Erwin P. Bottinger
    •  & Ruth J. F. Loos
  59. National Heart and Lung Institute, Cardiovascular Sciences, Hammersmith Campus, Imperial College London, London, UK

    • James Scott
    • , Sian-Tsung Tan
    •  & Jaspal Singh Kooner
  60. Department of Genome Sciences, University of Washington School of Medicine, Seattle, Washington, USA

    • Joshua D. Smith
  61. Analytic and Translational Genetics Unit, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts, USA

    • Benjamin M. Neale
    •  & Mark J. Daly
  62. Center for Genomic Medicine, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts, USA

    • Shaun Purcell
    •  & Jose C. Florez
  63. Department of Psychiatry, Icahn Institute for Genomics and Multiscale Biology, Icahn School of Medicine at Mount Sinai, New York, USA

    • Shaun Purcell
  64. Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK

    • Adam S. Butterworth
    • , Joanna M. M. Howson
    • , John Danesh
    •  & Manjinder Sandhu
  65. Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong, China

    • Heung Man Lee
    • , Vincent K. L. Lam
    • , Wing Yee So
    • , Claudia H. T. Tam
    • , Juliana CN Chan
    •  & Ronald C. W. Ma
  66. Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea

    • Soo-Heon Kwak
  67. Department of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA

    • Wei Zhao
  68. NIHR Blood and Transplant Research Unit in Donor Health and Genomics, Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK

    • John Danesh
  69. Department of Molecular Medicine and Biopharmaceutical Sciences, Graduate School of Convergence Science and Technology, and College of Medicine, Seoul National University, Seoul, Republic of Korea

    • Kyong Soo Park
  70. Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia, Pennsylvania, USA

    • Danish Saleheen
  71. Center for Non-Communicable Diseases, Karachi, Pakistan

    • Danish Saleheen
  72. Cardiovascular Division, Baylor College of Medicine, Houston, Texas, USA

    • David Aguilar
  73. Department of Pediatrics, University of Texas Health Science Center, San Antonio, Texas, USA

    • Rector Arya
    •  & Daniel Esten Hale
  74. Department of Medicine, Yong Loo Lin School of Medicine, National University of Singapore, National University Health System, Singapore, Singapore

    • Edmund Chan
    •  & E. Shyong Tai
  75. Department of Epidemiology, Murcia Regional Health Council, IMIB-Arrixaca, Murcia, Spain

    • Carmen Navarro
  76. CIBER Epidemiología y Salud Pública (CIBERESP), Spain

    • Carmen Navarro
  77. Unit of Preventive Medicine and Public Health, School of Medicine, University of Murcia, Spain

    • Carmen Navarro
  78. Cancer Research and Prevention Institute (ISPO), Florence, Italy

    • Domenico Palli
  79. Department of Medicine, University of Mississippi Medical Center, Jackson, Mississippi, USA

    • Adolfo Correa
    •  & Herman A. Taylor
  80. South Texas Diabetes and Obesity Institute, Regional Academic Health Center, University of Texas Health Science Center at San Antonio/University of Texas Rio Grande Valley, Brownsville, Texas, USA

    • Joanne E. Curran
    • , Satish Kumar
    •  & John Blangero
  81. Department of Genetics, Texas Biomedical Research Institute, San Antonio, Texas, USA

    • Vidya S. Farook
    • , Sobha Puppala
    •  & Ravindranath Duggirala
  82. Department of Internal Medicine, Section on Nephrology, Wake Forest School of Medicine, Winston-Salem, North Carolina, USA

    • Barry I. Freedman
  83. Center of Biostatistics and Bioinformatics, University of Mississippi Medical Center, Jackson, Mississippi, USA

    • Michael Griswold
  84. Department of Paediatrics, Yong Loo Lin School of Medicine, National University of Singapore, National University Health System, Singapore, Singapore

    • Chiea-Chuen Khor
  85. Division of Human Genetics, Genome Institute of Singapore, A*STAR, Singapore, Singapore

    • Chiea-Chuen Khor
    •  & Jianjun Liu
  86. CNRS-UMR8199, Lille University, Lille Pasteur Institute, Lille, France

    • Dorothée Thuillier
    • , Loïc Yengo
    •  & Philippe Froguel
  87. Institute of Health Sciences, University of Oulu, Oulu, Finland

    • Marie Loh
  88. Translational Laboratory in Genetic Medicine (TLGM), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore

    • Marie Loh
  89. Jackson Heart Study, University of Mississippi Medical Center, Jackson, Mississippi, USA

    • Solomon K. Musani
  90. College of Public Services, Jackson State University, Jackson, Mississippi, USA

    • Gregory Wilson
  91. KG Jebsen Center for Diabetes Research, Department of Clinical Science, University of Bergen, Bergen, Norway

    • Pål Rasmus Njølstad
  92. Department of Pediatrics, Haukeland University Hospital, Bergen, Norway

    • Pål Rasmus Njølstad
  93. NIHR Biomedical Research Centre at Guy’s and St Thomas’ Foundation Trust, London, UK

    • Massimo Mangino
  94. Institute of Human Genetics, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany

    • Thomas Schwarzmayr
    • , Thomas Wieland
    • , Tim M. Strom
    •  & Thomas Meitinger
  95. Department of Clinical Sciences, Diabetes and Endocrinology, Lund University Diabetes Centre, Malmö, Sweden

    • João Fadista
    • , Jasmina Kravic
    • , Valeri Lyssenko
    • , Claes Ladenvall
    • , Anders H. Rosengren
    •  & Leif Groop
  96. Institute of Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich Heine University, Düsseldorf, Germany

    • Christian Herder
    •  & Michael Roden
  97. German Center for Diabetes Research (DZD), München-Neuherberg, Germany

    • Christian Herder
    • , Jennifer Kriebel
    • , Wolfgang Rathmann
    • , Michael Roden
    • , Martin Hrabé de Angelis
    • , Barbara Thorand
    • , Christa Meisinger
    • , Annette Peters
    • , Cornelia Huth
    •  & Harald Grallert
  98. Institute of Regional Health Research, University of Southern Denmark, Odense, Denmark

    • Ivan Brandslund
  99. Department of Clinical Biochemistry, Vejle Hospital, Vejle, Denmark

    • Ivan Brandslund
  100. Department of Internal Medicine and Endocrinology, Vejle Hospital, Vejle, Denmark

    • Cramer Christensen
  101. Department of Health, National Institute for Health and Welfare, Helsinki, Finland

    • Heikki A. Koistinen
    •  & Leena Kinnunen
  102. Abdominal Center: Endocrinology, University of Helsinki and Helsinki University Central Hospital, Helsinki, Finland

    • Heikki A. Koistinen
    • , Liisa Hakaste
    •  & Tiinamaija Tuomi
  103. Minerva Foundation Institute for Medical Research, Helsinki, Finland

    • Heikki A. Koistinen
  104. Department of Medicine, University of Helsinki and Helsinki University Central Hospital, Helsinki, Finland

    • Heikki A. Koistinen
  105. Division of Cardiovascular and Diabetes Medicine, Medical Research Institute, Ninewells Hospital and Medical School, Dundee, UK

    • Alex S. F. Doney
  106. Estonian Genome Center, University of Tartu, Tartu, Estonia

    • Tõnu Esko
    • , Lili Milani
    • , Evelin Mihailov
    • , Andres Metspalu
    • , Reedik Mägi
    •  & Andrew P. Morris
  107. Department of Genetics, Harvard Medical School, Boston, Massachusetts, USA

    • Tõnu Esko
    •  & David Altshuler
  108. Division of Endocrinology, Boston Children's Hospital, Boston, Massachusetts, USA

    • Tõnu Esko
  109. Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK

    • Andrew J. Farmer
  110. Folkhälsan Research Centre, Helsinki, Finland

    • Liisa Hakaste
    • , Bo Isomaa
    •  & Tiinamaija Tuomi
  111. Research Programs Unit, Diabetes and Obesity, University of Helsinki, Helsinki, Finland

    • Liisa Hakaste
    •  & Tiinamaija Tuomi
  112. Steno Diabetes Center, Gentofte, Denmark

    • Marit E. Jørgensen
  113. Research Centre for Prevention and Health, Capital Region of Denmark, Glostrup, Denmark

    • Torben Jørgensen
    •  & Allan Linneberg
  114. Department of Public Health, Institute of Health Sciences, University of Copenhagen, Copenhagen, Denmark

    • Torben Jørgensen
  115. Faculty of Medicine, Aalborg University, Aalborg, Denmark

    • Torben Jørgensen
  116. Department of Primary Health Care, Vaasa Central Hospital, Vaasa, Finland

    • Annemari Käräjämäki
  117. Diabetes Center, Vaasa Health Care Center, Vaasa, Finland

    • Annemari Käräjämäki
  118. Institute of Epidemiology II, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany

    • Jennifer Kriebel
    • , Barbara Thorand
    • , Christa Meisinger
    • , Annette Peters
    • , Cornelia Huth
    • , Harald Grallert
    •  & Christian Gieger
  119. Research Unit of Molecular Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany

    • Jennifer Kriebel
    • , Harald Grallert
    • , Christian Gieger
    •  & Thomas Illig
  120. Institute for Biometrics and Epidemiology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich Heine University, Düsseldorf, Germany

    • Wolfgang Rathmann
  121. Department of Public Health, Section of General Practice, Aarhus University, Aarhus, Denmark

    • Torsten Lauritzen
  122. Department of Clinical Experimental Research, Rigshospitalet, Glostrup, Denmark

    • Allan Linneberg
  123. Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

    • Allan Linneberg
  124. Department of Clinical Sciences, Hypertension and Cardiovascular Disease, Lund University, Malmö, Sweden

    • Olle Melander
  125. Oxford NIHR Biomedical Research Centre, Oxford University Hospitals Trust, Oxford, UK

    • Matt Neville
    • , Fredrik Karpe
    • , Katharine R. Owen
    • , Anna L. Gloyn
    •  & Mark I. McCarthy
  126. Department of Clinical Sciences, Diabetes and Cardiovascular Disease, Genetic Epidemiology, Lund University, Malmö, Sweden

    • Marju Orho-Melander
  127. Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts, USA

    • Lu Qi
    • , Qibin Qi
    • , Frank B. Hu
    •  & Paul W. Franks
  128. Channing Division of Network Medicine, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts, USA

    • Lu Qi
  129. Department of Epidemiology and Population Health, Albert Einstein College of Medicine, New York, USA

    • Qibin Qi
  130. Division of Endocrinology and Diabetology, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany

    • Michael Roden
  131. Department of Public Health and Clinical Medicine, Umeå University, Umeå, Sweden

    • Olov Rolandsson
    •  & Paul W. Franks
  132. High Throughput Genomics, Oxford Genomics Centre, Wellcome Trust Centre for Human Genetics, Nuffield Department of Medicine, University of Oxford, Oxford, UK

    • Christine Blancher
    • , Gemma Buck
    • , Joseph Trakalo
    •  & David Buck
  133. Institute of Experimental Genetics, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany

    • Martin Hrabé de Angelis
  134. Center of Life and Food Sciences Weihenstephan, Technische Universität München, Freising-Weihenstephan, Germany

    • Martin Hrabé de Angelis
  135. William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK

    • Panos Deloukas
  136. Princess Al-Jawhara Al-Brahim Centre of Excellence in Research of Hereditary Disorders (PACER-HD), King Abdulaziz University, Jeddah, Saudi Arabia

    • Panos Deloukas
  137. Department of Clinical Sciences, Medicine, Lund University, Malmö, Sweden

    • Peter Nilsson
  138. Faculty of Health Sciences, University of Southern Denmark, Odense, Denmark

    • Torben Hansen
  139. Department of Social Services and Health Care, Jakobstad, Finland

    • Bo Isomaa
  140. Metabolic Research Laboratories, Institute of Metabolic Science, University of Cambridge, Cambridge, UK

    • Stephen P O'Rahilly
    •  & Inês Barroso
  141. Pat Macpherson Centre for Pharmacogenetics and Pharmacogenomics, Medical Research Institute, Ninewells Hospital and Medical School, Dundee, UK

    • Colin N. A. Palmer
  142. Foundation for Research in Health, Exercise and Nutrition, Kuopio Research Institute of Exercise Medicine, Kuopio, Finland

    • Rainer Rauramaa
  143. Center for Vascular Prevention, Danube University Krems, Krems, Austria

    • Jaakko Tuomilehto
  144. Diabetes Research Group, King Abdulaziz University, Jeddah, Saudi Arabia

    • Jaakko Tuomilehto
  145. Dasman Diabetes Institute, Dasman, Kuwait

    • Jaakko Tuomilehto
  146. National Institute for Health and Welfare, Helsinki, Finland

    • Jaakko Tuomilehto
    •  & Veikko Salomaa
  147. Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, California, USA

    • Richard M. Watanabe
  148. Department of Physiology & Biophysics, Keck School of Medicine, University of Southern California, Los Angeles, California, USA

    • Richard M. Watanabe
  149. Diabetes and Obesity Research Institute, Keck School of Medicine, University of Southern California, Los Angeles, California, USA

    • Richard M. Watanabe
  150. Department of Medical Sciences, Molecular Medicine and Science for Life Laboratory, Uppsala University, Uppsala, Sweden

    • Ann-Christine Syvänen
  151. Cedars-Sinai Diabetes and Obesity Research Institute, Los Angeles, California, USA

    • Richard N. Bergman
  152. Functional Genomics Unit, CSIR-Institute of Genomics & Integrative Biology (CSIR-IGIB), New Delhi, India

    • Dwaipayan Bharadwaj
  153. Department of Biomedical Science, Hallym University, Chuncheon, Republic of Korea

    • Yoon Shin Cho
  154. CSIR-Centre for Cellular and Molecular Biology, Hyderabad, Telangana, India

    • Giriraj R. Chandak
  155. Li Ka Shing Institute of Health Sciences, The Chinese University of Hong Kong, Hong Kong, China

    • Juliana CN Chan
    •  & Ronald C. W. Ma
  156. Hong Kong Institute of Diabetes and Obesity, The Chinese University of Hong Kong, Hong Kong, China

    • Juliana CN Chan
    •  & Ronald C. W. Ma
  157. MRC-PHE Centre for Environment and Health, Imperial College London, London, UK

    • Paul Elliott
  158. The Biostatistics Center, The George Washington University, Rockville, Maryland, USA

    • Kathleen A. Jablonski
  159. Department of Medicine, Division of Endocrinology, Diabetes and Nutrition, and Program in Personalized and Genomic Medicine, University of Maryland School of Medicine, Baltimore, Maryland, USA

    • Toni I. Pollin
  160. Department of Endocrinology and Metabolism, All India Institute of Medical Sciences, New Delhi, India

    • Nikhil Tandon
  161. Department of Genomics of Common Disease, School of Public Health, Imperial College London, London, UK

    • Philippe Froguel
    •  & Inga Prokopenko
  162. Life Sciences Institute, National University of Singapore, Singapore, Singapore

    • Yik Ying Teo
  163. Department of Statistics and Applied Probability, National University of Singapore, Singapore, Singapore

    • Yik Ying Teo
  164. Endocrinology and Metabolism Service, Hadassah-Hebrew University Medical Center, Jerusalem, Israel

    • Benjamin Glaser
  165. The Medical School, Institute of Cellular Medicine, Newcastle University, Newcastle, UK

    • Mark Walker
  166. Department of Medical Sciences, Molecular Epidemiology and Science for Life Laboratory, Uppsala University, Uppsala, Sweden

    • Erik Ingelsson
  167. Hannover Unified Biobank, Hannover Medical School, Hanover, Germany

    • Thomas Illig
  168. Department of Human Genetics, Hannover Medical School, Hanover, Germany

    • Thomas Illig
  169. Department of Medical Sciences, Uppsala University, Uppsala, Sweden

    • Lars Lind
  170. Data Sciences and Data Engineering, Broad Institute, Cambridge, Massachusetts, USA

    • Yossi Farjoun
  171. Finnish Institute for Molecular Medicine, University of Helsinki, Helsinki, Finland

    • Tiinamaija Tuomi
    •  & Leif Groop
  172. Imperial College Healthcare NHS Trust, Imperial College London, London, UK

    • Jaspal Singh Kooner
    •  & John C. Chambers
  173. Clinical Research Centre, Centre for Molecular Medicine, Ninewells Hospital and Medical School, Dundee, UK

    • Andrew D. Morris
  174. The Usher Institute to the Population Health Sciences and Informatics, University of Edinburgh, Edinburgh, UK

    • Andrew D. Morris
  175. University of Exeter Medical School, University of Exeter, Exeter, UK

    • Andrew T. Hattersley
  176. Department of Natural Science, University of Haifa, Haifa, Israel

    • Gil Atzmon
  177. Institute of Human Genetics, Technische Universität München, Munich, Germany

    • Tim M. Strom
    •  & Thomas Meitinger
  178. Departments of Medicine and Human Genetics, The University of Chicago, Chicago, Illinois, USA

    • Graeme I. Bell
  179. Cardiovascular & Metabolic Disorders Program, Duke-NUS Medical School Singapore, Singapore, Singapore

    • E. Shyong Tai
  180. Li Ka Shing Centre for Health Information and Discovery, University of Oxford, Oxford, UK

    • Gilean McVean
  181. Department of Physiology and Biophysics, University of Mississippi Medical Center, Jackson, Mississippi, USA

    • James G. Wilson
  182. Department of Laboratory Medicine & Institute for Human Genetics, University of California, San Francisco, San Francisco, California, USA

    • Mark Seielstad
  183. Blood Systems Research Institute, San Francisco, California, USA

    • Mark Seielstad
  184. General Medicine Division, Massachusetts General Hospital and Department of Medicine, Harvard Medical School, Boston, Massachusetts, USA

    • James B. Meigs
  185. Division of Endocrinology and Metabolism, Department of Medicine, McGill University, Montreal, Quebec, Canada

    • Rob Sladek
  186. Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA

    • Eric S. Lander
  187. Department of Genetics, University of North Carolina, Chapel Hill, North Carolina, USA

    • Karen L. Mohlke
  188. Department of Biostatistics, University of Liverpool, Liverpool, UK

    • Andrew P. Morris
  189. Department of Medicine, Harvard Medical School, Boston, Massachusetts, USA

    • David Altshuler
    •  & Jose C. Florez
  190. Diabetes Research Center (Diabetes Unit), Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts, USA

    • David Altshuler
    •  & Jose C. Florez
  191. Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA

    • David Altshuler

Authors

  1. Search for Jason Flannick in:

  2. Search for Christian Fuchsberger in:

  3. Search for Anubha Mahajan in:

  4. Search for Tanya M. Teslovich in:

  5. Search for Vineeta Agarwala in:

  6. Search for Kyle J. Gaulton in:

  7. Search for Lizz Caulkins in:

  8. Search for Ryan Koesterer in:

  9. Search for Clement Ma in:

  10. Search for Loukas Moutsianas in:

  11. Search for Davis J. McCarthy in:

  12. Search for Manuel A. Rivas in:

  13. Search for John R. B. Perry in:

  14. Search for Xueling Sim in:

  15. Search for Thomas W. Blackwell in:

  16. Search for Neil R. Robertson in:

  17. Search for N William Rayner in:

  18. Search for Pablo Cingolani in:

  19. Search for Adam E. Locke in:

  20. Search for Juan Fernandez Tajes in:

  21. Search for Heather M. Highland in:

  22. Search for Josee Dupuis in:

  23. Search for Peter S. Chines in:

  24. Search for Cecilia M. Lindgren in:

  25. Search for Christopher Hartl in:

  26. Search for Anne U. Jackson in:

  27. Search for Han Chen in:

  28. Search for Jeroen R. Huyghe in:

  29. Search for Martijn van de Bunt in:

  30. Search for Richard D. Pearson in:

  31. Search for Ashish Kumar in:

  32. Search for Martina Müller-Nurasyid in:

  33. Search for Niels Grarup in:

  34. Search for Heather M. Stringham in:

  35. Search for Eric R. Gamazon in:

  36. Search for Jaehoon Lee in:

  37. Search for Yuhui Chen in:

  38. Search for Robert A. Scott in:

  39. Search for Jennifer E. Below in:

  40. Search for Peng Chen in:

  41. Search for Jinyan Huang in:

  42. Search for Min Jin Go in:

  43. Search for Michael L. Stitzel in:

  44. Search for Dorota Pasko in:

  45. Search for Stephen C. J. Parker in:

  46. Search for Tibor V. Varga in:

  47. Search for Todd Green in:

  48. Search for Nicola L. Beer in:

  49. Search for Aaron G. Day-Williams in:

  50. Search for Teresa Ferreira in:

  51. Search for Tasha Fingerlin in:

  52. Search for Momoko Horikoshi in:

  53. Search for Cheng Hu in:

  54. Search for Iksoo Huh in:

  55. Search for Mohammad Kamran Ikram in:

  56. Search for Bong-Jo Kim in:

  57. Search for Yongkang Kim in:

  58. Search for Young Jin Kim in:

  59. Search for Min-Seok Kwon in:

  60. Search for Juyoung Lee in:

  61. Search for Selyeong Lee in:

  62. Search for Keng-Han Lin in:

  63. Search for Taylor J. Maxwell in:

  64. Search for Yoshihiko Nagai in:

  65. Search for Xu Wang in:

  66. Search for Ryan P. Welch in:

  67. Search for Joon Yoon in:

  68. Search for Weihua Zhang in:

  69. Search for Nir Barzilai in:

  70. Search for Benjamin F. Voight in:

  71. Search for Bok-Ghee Han in:

  72. Search for Christopher P. Jenkinson in:

  73. Search for Teemu Kuulasmaa in:

  74. Search for Johanna Kuusisto in:

  75. Search for Alisa Manning in:

  76. Search for Maggie C. Y. Ng in:

  77. Search for Nicholette D. Palmer in:

  78. Search for Beverley Balkau in:

  79. Search for Alena Stančáková in:

  80. Search for Hanna E. Abboud in:

  81. Search for Heiner Boeing in:

  82. Search for Vilmantas Giedraitis in:

  83. Search for Dorairaj Prabhakaran in:

  84. Search for Omri Gottesman in:

  85. Search for James Scott in:

  86. Search for Jason Carey in:

  87. Search for Phoenix Kwan in:

  88. Search for George Grant in:

  89. Search for Joshua D. Smith in:

  90. Search for Benjamin M. Neale in:

  91. Search for Shaun Purcell in:

  92. Search for Adam S. Butterworth in:

  93. Search for Joanna M. M. Howson in:

  94. Search for Heung Man Lee in:

  95. Search for Yingchang Lu in:

  96. Search for Soo-Heon Kwak in:

  97. Search for Wei Zhao in:

  98. Search for John Danesh in:

  99. Search for Vincent K. L. Lam in:

  100. Search for Kyong Soo Park in:

  101. Search for Danish Saleheen in:

  102. Search for Wing Yee So in:

  103. Search for Claudia H. T. Tam in:

  104. Search for Uzma Afzal in:

  105. Search for David Aguilar in:

  106. Search for Rector Arya in:

  107. Search for Tin Aung in:

  108. Search for Edmund Chan in:

  109. Search for Carmen Navarro in:

  110. Search for Ching-Yu Cheng in:

  111. Search for Domenico Palli in:

  112. Search for Adolfo Correa in:

  113. Search for Joanne E. Curran in:

  114. Search for Dennis Rybin in:

  115. Search for Vidya S. Farook in:

  116. Search for Sharon P. Fowler in:

  117. Search for Barry I. Freedman in:

  118. Search for Michael Griswold in:

  119. Search for Daniel Esten Hale in:

  120. Search for Pamela J. Hicks in:

  121. Search for Chiea-Chuen Khor in:

  122. Search for Satish Kumar in:

  123. Search for Benjamin Lehne in:

  124. Search for Dorothée Thuillier in:

  125. Search for Wei Yen Lim in:

  126. Search for Jianjun Liu in:

  127. Search for Marie Loh in:

  128. Search for Solomon K. Musani in:

  129. Search for Sobha Puppala in:

  130. Search for William R. Scott in:

  131. Search for Loïc Yengo in:

  132. Search for Sian-Tsung Tan in:

  133. Search for Herman A. Taylor in:

  134. Search for Farook Thameem in:

  135. Search for Gregory Wilson in:

  136. Search for Tien Yin Wong in:

  137. Search for Pål Rasmus Njølstad in:

  138. Search for Jonathan C. Levy in:

  139. Search for Massimo Mangino in:

  140. Search for Lori L. Bonnycastle in:

  141. Search for Thomas Schwarzmayr in:

  142. Search for João Fadista in:

  143. Search for Gabriela L. Surdulescu in:

  144. Search for Christian Herder in:

  145. Search for Christopher J. Groves in:

  146. Search for Thomas Wieland in:

  147. Search for Jette Bork-Jensen in:

  148. Search for Ivan Brandslund in:

  149. Search for Cramer Christensen in:

  150. Search for Heikki A. Koistinen in:

  151. Search for Alex S. F. Doney in:

  152. Search for Leena Kinnunen in:

  153. Search for Tõnu Esko in:

  154. Search for Andrew J. Farmer in:

  155. Search for Liisa Hakaste in:

  156. Search for Dylan Hodgkiss in:

  157. Search for Jasmina Kravic in:

  158. Search for Valeri Lyssenko in:

  159. Search for Mette Hollensted in:

  160. Search for Marit E. Jørgensen in:

  161. Search for Torben Jørgensen in:

  162. Search for Claes Ladenvall in:

  163. Search for Johanne Marie Justesen in:

  164. Search for Annemari Käräjämäki in:

  165. Search for Jennifer Kriebel in:

  166. Search for Wolfgang Rathmann in:

  167. Search for Lars Lannfelt in:

  168. Search for Torsten Lauritzen in:

  169. Search for Narisu Narisu in:

  170. Search for Allan Linneberg in:

  171. Search for Olle Melander in:

  172. Search for Lili Milani in:

  173. Search for Matt Neville in:

  174. Search for Marju Orho-Melander in:

  175. Search for Lu Qi in:

  176. Search for Qibin Qi in:

  177. Search for Michael Roden in:

  178. Search for Olov Rolandsson in:

  179. Search for Amy Swift in:

  180. Search for Anders H. Rosengren in:

  181. Search for Kathleen Stirrups in:

  182. Search for Andrew R. Wood in:

  183. Search for Evelin Mihailov in:

  184. Search for Christine Blancher in:

  185. Search for Mauricio O. Carneiro in:

  186. Search for Jared Maguire in:

  187. Search for Ryan Poplin in:

  188. Search for Khalid Shakir in:

  189. Search for Timothy Fennell in:

  190. Search for Mark DePristo in:

  191. Search for Martin Hrabé de Angelis in:

  192. Search for Panos Deloukas in:

  193. Search for Anette P. Gjesing in:

  194. Search for Goo Jun in:

  195. Search for Peter Nilsson in:

  196. Search for Jacquelyn Murphy in:

  197. Search for Robert Onofrio in:

  198. Search for Barbara Thorand in:

  199. Search for Torben Hansen in:

  200. Search for Christa Meisinger in:

  201. Search for Frank B. Hu in:

  202. Search for Bo Isomaa in:

  203. Search for Fredrik Karpe in:

  204. Search for Liming Liang in:

  205. Search for Annette Peters in:

  206. Search for Cornelia Huth in:

  207. Search for Stephen P O'Rahilly in:

  208. Search for Colin N. A. Palmer in:

  209. Search for Oluf Pedersen in:

  210. Search for Rainer Rauramaa in:

  211. Search for Jaakko Tuomilehto in:

  212. Search for Veikko Salomaa in:

  213. Search for Richard M. Watanabe in:

  214. Search for Ann-Christine Syvänen in:

  215. Search for Richard N. Bergman in:

  216. Search for Dwaipayan Bharadwaj in:

  217. Search for Erwin P. Bottinger in:

  218. Search for Yoon Shin Cho in:

  219. Search for Giriraj R. Chandak in:

  220. Search for Juliana CN Chan in:

  221. Search for Kee Seng Chia in:

  222. Search for Mark J. Daly in:

  223. Search for Shah B. Ebrahim in:

  224. Search for Claudia Langenberg in:

  225. Search for Paul Elliott in:

  226. Search for Kathleen A. Jablonski in:

  227. Search for Donna M. Lehman in:

  228. Search for Weiping Jia in:

  229. Search for Ronald C. W. Ma in:

  230. Search for Toni I. Pollin in:

  231. Search for Manjinder Sandhu in:

  232. Search for Nikhil Tandon in:

  233. Search for Philippe Froguel in:

  234. Search for Inês Barroso in:

  235. Search for Yik Ying Teo in:

  236. Search for Eleftheria Zeggini in:

  237. Search for Ruth J. F. Loos in:

  238. Search for Kerrin S. Small in:

  239. Search for Janina S. Ried in:

  240. Search for Ralph A. DeFronzo in:

  241. Search for Harald Grallert in:

  242. Search for Benjamin Glaser in:

  243. Search for Andres Metspalu in:

  244. Search for Nicholas J. Wareham in:

  245. Search for Mark Walker in:

  246. Search for Eric Banks in:

  247. Search for Christian Gieger in:

  248. Search for Erik Ingelsson in:

  249. Search for Hae Kyung Im in:

  250. Search for Thomas Illig in:

  251. Search for Paul W. Franks in:

  252. Search for Gemma Buck in:

  253. Search for Joseph Trakalo in:

  254. Search for David Buck in:

  255. Search for Inga Prokopenko in:

  256. Search for Reedik Mägi in:

  257. Search for Lars Lind in:

  258. Search for Yossi Farjoun in:

  259. Search for Katharine R. Owen in:

  260. Search for Anna L. Gloyn in:

  261. Search for Konstantin Strauch in:

  262. Search for Tiinamaija Tuomi in:

  263. Search for Jaspal Singh Kooner in:

  264. Search for Jong-Young Lee in:

  265. Search for Taesung Park in:

  266. Search for Peter Donnelly in:

  267. Search for Andrew D. Morris in:

  268. Search for Andrew T. Hattersley in:

  269. Search for Donald W. Bowden in:

  270. Search for Francis S. Collins in:

  271. Search for Gil Atzmon in:

  272. Search for John C. Chambers in:

  273. Search for Timothy D. Spector in:

  274. Search for Markku Laakso in:

  275. Search for Tim M. Strom in:

  276. Search for Graeme I. Bell in:

  277. Search for John Blangero in:

  278. Search for Ravindranath Duggirala in:

  279. Search for E. Shyong Tai in:

  280. Search for Gilean McVean in:

  281. Search for Craig L. Hanis in:

  282. Search for James G. Wilson in:

  283. Search for Mark Seielstad in:

  284. Search for Timothy M. Frayling in:

  285. Search for James B. Meigs in:

  286. Search for Nancy J. Cox in:

  287. Search for Rob Sladek in:

  288. Search for Eric S. Lander in:

  289. Search for Stacey Gabriel in:

  290. Search for Karen L. Mohlke in:

  291. Search for Thomas Meitinger in:

  292. Search for Leif Groop in:

  293. Search for Goncalo Abecasis in:

  294. Search for Laura J. Scott in:

  295. Search for Andrew P. Morris in:

  296. Search for Hyun Min Kang in:

  297. Search for David Altshuler in:

  298. Search for Noël P. Burtt in:

  299. Search for Jose C. Florez in:

  300. Search for Michael Boehnke in:

  301. Search for Mark I. McCarthy in:

Contributions

Author contributions are described in the Supplementary Information.

Competing interests

Ralph A DeFronzo has been a member of advisory boards for Astra Zeneca, Novo Nordisk, Janssen, Lexicon, Boehringer-Ingelheim, received research support from Bristol Myers Squibb, Boehringer- Ingelheim, Takeda and Astra Zeneca, and is a member of speaker’s bureaus for Novo-Nordisk and Astra Zeneca. Jose C Florez has received consulting honoraria from Pfizer and PanGenX. Erik Ingelsson is an advisor and consultant for Precision Wellness, Inc., and advisor for Cellink for work unrelated to the present project. Mark McCarthy has received consulting and advisory board honoraria from Pfizer, Lilly, and NovoNordisk. Gilean McVean and Peter Donnelly are co-founders of Genomics PLC, which provides genome analytics.

Corresponding author

Correspondence to Jason Flannick.

Supplementary information

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/sdata.2017.179