Data available for download

We are releasing the summary data from our genome-wide meta analyses of glycaemic traits, in order to enable other researchers to examine particular variants or loci of their interest for association with these traits. The files include p-values and direction of effect at over 2 million directly genotyped or imputed single nucleotide polymorphisms (SNPs), as well as frequency information from the HapMap project (release 27).

Acknowleding the data

When using data from the downloadable meta-analyses results please acknowledge the source of the data as follows: "Data on glycaemic traits have been contributed by MAGIC investigators and have been downloaded from www.magicinvestigators.org".

In addition to the above acknowledgement, please cite the relevant paper.

Downloading the data

The data can be downloaded from the magic directory on the Exeter FTP site:

• MAGIC FTP

Datasets

Genome-wide association study summary statistics of random glucose in 476,326 individuals

Random Glucose European, Trans-Ethnic meta-analyses, sex-specific and sex-dimorphic UKBB & Vanderbilt meta-analyses public data release. The summary statistics we provide for the European and Trans-Ethnic meta-analyses were filtered to exclude non-biallelic variants, indels, variants available in less than 50% of the total sample and variants with MAF < 0.01. Further information and file format details are provided in the accompanying README.

Please cite:

• GWAS of random glucose in 476,326 individuals provide insights into diabetes pathophysiology, complications and treatment stratification

  Vasiliki Lagou et al.

  Nature Genetics 2023

  PUBMED: 37679419; DOI: 10.1038/s41588-023-01462-3

The following summary statistics can be downloaded:

• MAGIC_RG_MA_EUR_2019May31.txt.gz
• MAGIC_RG_MA_transEthnic_2020Oct08.txt.gz
• MAGIC_RG_MA_SexSpecific_2020.txt

Genome-wide association study summary statistics for insulin-stimulated glucose uptake

These files include summary results for the Modified Stumoll ISI and insulin fold change meta-analyses presented in Williamson et al (2023). A total of 55,535 and 55,172 participants without diabetes across 28 studies were included in analyses of Modified Stumvoll ISI and insulin fold change, respectively. Two of these studies included participants of Hispanic American ancestry (MACAD and HTN-IR) and 1 included participants of East Asian ancestry (TAICHI); the remaining 25 studies included individuals of European ancestry. Further information and file format details are provided in the accompanying README.

Please cite:

• Genome-wide association study and functional characterisation identifies candidate genes for insulin-stimulated glucose uptake

  Williamson et al.

  Nature Genetics 2023

  PUBMED: ; DOI: 10.1038/s41588-023-01408-9

The following summary statistics can be downloaded:

• IFC_adjBMI_ALL.tsv.gz
• IFC_adjBMI_EUR.tsv.gz
• IFC_adjBMI_NONEUR.tsv.gz
• IFC_noBMI_ALL.tsv.gz
• IFC_noBMI_EUR.tsv.gz
• IFC_noBMI_NONEUR.tsv.gz
• ISI_adjBMI_ALL.tsv.gz
• ISI_adjBMI_EUR.tsv.gz
• ISI_adjBMI_NONEUR.tsv.gz
• ISI_noBMI_ALL.tsv.gz
• ISI_noBMI_EUR.tsv.gz
• ISI_noBMI_NONEUR.tsv.gz

Loci for insulin processing and secretion provide insight into type 2 diabetes risk

These files include summary results for the proinsulin meta-analysis (in press, Broadaway et al 2023 AJHG). The meta-analysis was composed of 16 European-ancestry studies. The model adjusting for BMI had maximum sample size of 45,826, while the model not adjusting for BMI had a maximum sample size of 45,861 individuals. Further information and file format details are provided in the accompanying README.

Please cite:

• Loci for insulin processing and secretion provide insight into type 2 diabetes risk

  Kelsy Alaine Broadway et al.

   

  PUBMED: 36693378; DOI: 10.1016/j.ajhg.2023.01.002

The following summary statistics can be downloaded:

• PROIadjBMI_META_GWAS_CAT_20221113.tsv.gz
• PROInoBMI_META_GWAS_CAT_20221113.tsv.gz

Large-scale exome array summary statistics resources for glycemic traits to aid effector gene prioritization

Multi-ancestry meta-analysis and single-ancestry meta-analysis results using raremetalworker of up to 144,060 individuals without diabetes with fasting glucose (FG), 2h-glucose post-challenge (2hGlu), glycated haemoglobin (HbA1c), and fasting insulin data (FI). Please note that for simplicity in the manuscripts we have used the acronyms FG, FI and 2hGlu however all these traits and associated results have been adjusted for BMI. Further information and file format details are provided in the accompanying README.

Please cite:

• Large-scale exome array summary statistics resources for glycemic traits to aid effector gene prioritization

  Sara Willems et al.

   

The following summary statistics can be downloaded:

• MAGICExome_2hGlu_ALL.tsv.gz
• MAGICExome_2hGlu_AFR.tsv.gz
• MAGICExome_2hGlu_EUR.tsv.gz
• MAGICExome_FG_ALL.tsv.gz
• MAGICExome_FG_AFR.tsv.gz
• MAGICExome_FG_EAS.tsv.gz
• MAGICExome_FG_EUR.tsv.gz
• MAGICExome_FG_HISP.tsv.gz
• MAGICExome_FG_SAS.tsv.gz
• MAGICExome_FI_ALL.tsv.gz
• MAGICExome_FI_AFR.tsv.gz
• MAGICExome_FI_EAS.tsv.gz
• MAGICExome_FI_EUR.tsv.gz
• MAGICExome_FI_HISP.tsv.gz
• MAGICExome_FI_SAS.tsv.gz
• MAGICExome_HbA1c_ALL.tsv.gz
• MAGICExome_HbA1c_AFR.tsv.gz
• MAGICExome_HbA1c_EAS.tsv.gz
• MAGICExome_HbA1c_EUR.tsv.gz
• MAGICExome_HbA1c_HISP.tsv.gz
• MAGICExome_HbA1c_SAS.tsv.gz

Trans-ancestry and single-ancestry meta-analysis summary statistics of four glycaemic traits

Trans-ancestry meta-analysis results using MANTRA and single-ancestry meta-analysis results using METAL of up to 281,416 individuals without diabetes with fasting glucose (FG), 2h-glucose post-challenge (2hGlu), glycated haemoglobin (HbA1c), and fasting insulin data (FI). Please note that for simplicity in the manuscripts we have used the acronyms FG, FI and 2hGlu however all these traits and associated results have been adjusted for BMI. Further information and file format details are provided in the accompanying README.

Please cite:

• The Trans-Ancestral Genomic Architecture of Glycaemic Traits

  Ji Chen et al.

  Nature Genetics 2021

  PUBMED: 34059833; PMC: 7610958; DOI: 10.1038/s41588-021-00852-9

The following summary statistics can be downloaded:

• MAGIC1000G_2hGlu_TA.tsv.gz
• MAGIC1000G_2hGlu_AA.tsv.gz
• MAGIC1000G_2hGlu_EAS.tsv.gz
• MAGIC1000G_2hGlu_EUR.tsv.gz
• MAGIC1000G_2hGlu_HISP.tsv.gz
• MAGIC1000G_FG_TA.tsv.gz
• MAGIC1000G_FG_AA.tsv.gz
• MAGIC1000G_FG_EAS.tsv.gz
• MAGIC1000G_FG_EUR.tsv.gz
• MAGIC1000G_FG_HISP.tsv.gz
• MAGIC1000G_FG_SAS.tsv.gz
• MAGIC1000G_FI_TA.tsv.gz
• MAGIC1000G_FI_AA.tsv.gz
• MAGIC1000G_FI_EAS.tsv.gz
• MAGIC1000G_FI_EUR.tsv.gz
• MAGIC1000G_FI_HISP.tsv.gz
• MAGIC1000G_FI_SAS.tsv.gz
• MAGIC1000G_HbA1c_TA.tsv.gz
• MAGIC1000G_HbA1c_AA.tsv.gz
• MAGIC1000G_HbA1c_EAS.tsv.gz
• MAGIC1000G_HbA1c_EUR.tsv.gz
• MAGIC1000G_HbA1c_HISP.tsv.gz
• MAGIC1000G_HbA1c_SAS.tsv.gz
• MAGIC1000G_HbA1c_Uganda.tsv.gz

The following genetic scores can be downloaded:

• FG_genetic_score.xlsx
• HbA1c_genetic_score.xlsx
• FI_genetic_score.xlsx
• 2hGlu_genetic_score.xlsx
• FG_EUR_PGS.tsv.gz
• HbA1c_EUR_PGS.tsv.gz
• FI_EUR_PGS.tsv.gz

Fasting glucose and fasting insulin sex-specific and sex-differentiated GWAS meta-analysis summary statistics

Sex-specific summary statistics for up to 67,506 men / 73,089 women (FG meta-analysis) and 47,806 men / 50,404 women (FI meta-analysis). Further information and file format details are provided in the accompanying README.

Please cite:

• Sex-dimorphic genetic effects and novel loci for fasting glucose and insulin variability.

  Lagou V, Mägi R, Hottenga JJ et al.

  Nature Communications 2021;12;24;

  PUBMED: 33402679; PMC: 7785747; DOI: 10.1038/s41467-020-19366-9

The following data can be downloaded:

• FG_combined_1000G_density_formatted_21-03-29.txt.gz
• FG_female_1000G_density_formatted_21-03-29.txt.gz
• FG_male_1000G_density_formatted_21-03-29.txt.gz
• FI_combined_1000G_density_formatted_21-03-29.txt.gz
• FI_female_1000G_density_formatted_21-03-29.txt.gz
• FI_male_1000G_density_formatted_21-03-29.txt.gz
• FG_STAGE1_2_3_SEX_GWAS_updated_2019.txt.gz
• FI_STAGE1_2_3_SEX_GWAS_updated_2019.txt.gz

Change in Fasting Glucose Over Time

Results from a genome-wide association study of longitudinal fasting glucose changes in up to 13,807 non-diabetic individuals of European descent from nine cohorts. Fasting glucose change over time was defined as the slope of the line defined by multiple fasting glucose measurements obtained over up to 14 years of observation. File format details are provided in the accompanying README.

Please cite:

• Genome-wide Association Study of Change in Fasting Glucose over time in 13,807 non-diabetic European Ancestry Individuals.

  Liu CT, Merino J, Rybin D, DiCorpo D, Benke KS et al.

  Scientific reports 2019;9;1;9439

  PUBMED: 31263163; PMC: 6602949; DOI: 10.1038/s41598-019-45823-7

The following data can be downloaded:

• Longitudinal_2019_FGovertime_corrected_meta_filtered_ready.txt.gz
• Longitudinal_2019_README.txt

HbA1c summary statistics (ancestry-specific and transethnic)

Ancestry-specific and transethnic genome-wide meta-analysis summary statistics for association with HbA1c in up to 159,940 individuals from 82 cohorts of European (N=123,665), African (N=7,564), East Asian (N=20,838) and South Asian (N=8,874) ancestry. HbA1c trait values are untransformed and adjusted for age, sex and study-specific covariates. File format details are provided in the accompanying README.

Please cite:

• Impact of common genetic determinants of Hemoglobin A1c on type 2 diabetes risk and diagnosis in ancestrally diverse populations: A transethnic genome-wide meta-analysis.

  Wheeler E et al.

  PLoS medicine 2017;14;9;e1002383

  PUBMED: 28898252; PMC: 5595282; DOI: 10.1371/journal.pmed.1002383

The following data can be downloaded:

There was an error in the earlier versions of the African American and South Asian summary statistics and these files have been updated. Please replace earlier versions with the Sept2018 versions below. Apologies for any inconvenience.

• HbA1c_MANTRA.txt.gz
• *Updated* HbA1c_METAL_AfricanAmerican_updatedSept2018.txt.gz
• HbA1c_METAL_EastAsian.txt.gz
• HbA1c_METAL_European.txt.gz
• *Updated* HbA1c_METAL_SouthAsian_updatedSept2018.txt.gz
• README.txt

Modified Stumvoll Insulin Sensitivity Index (ISI) datasets

Meta-analysis results files for the modified Stumvoll Insulin Sensitivity Index (ISI). Discovery was performed in 16,753 individuals, and replication was attempted for the 23 most significant novel loci in 13,354 independent individuals. Data are provided for the discovery effort. Details of the models are within the in the README accompanying the data. Results of the combined discovery and replication efforts are available in the publication.

Please cite:

• Genome-Wide Association Study of the Modified Stumvoll Insulin Sensitivity Index Identifies BCL2 and FAM19A2 as Novel Insulin Sensitivity Loci.

  Walford GA et al.

  Diabetes 2016;65;10;3200-11

  PUBMED: 27416945; PMC: 5033262; DOI: 10.2337/db16-0199

The following data can be downloaded:

• MAGIC_ISI_Model_1_AgeSexOnly.txt
• MAGIC_ISI_Model_2_AgeSexBMI.txt
• MAGIC_ISI_Model_3_JMA.txt
• MAGIC_ISI_README.docx

Insulin secretion during OGTT

Meta-analysis results files for glucose-stimulated insulin secretion (GSIS) indices during an oral glucose tolerance test in up to 5318 non-diabetic participants from 9 cohorts.

Please cite:

• A central role for GRB10 in regulation of islet function in man.

  Prokopenko I et al.

  PLoS genetics 2014;10;4;e1004235

  PUBMED: 24699409; PMC: 3974640; DOI: 10.1371/journal.pgen.1004235

The results file and a README can be downloaded:

• MAGIC_InsulinSecretion_data_release_May2014.zip

Metabochip replication datasets

Results for fasting glucose are from models adjusted for age and sex, and from up to 133,010 non-diabetic participants from 66 studies. Fasting insulin results are for ln-transformed fasting insulin as the outcome and are adjusted for age, sex and are reported both with and without BMI adjustment. These results are from up to 108,557 individuals from 56 studies. Results for 2h-glucose are from models adjusted for age and sex and from up to 42,854 individuals from 20 studies.

Please cite:

• Large-scale association analyses identify new loci influencing glycemic traits and provide insight into the underlying biological pathways.

  Scott RA et al.

  Nature genetics 2012;44;9;991-1005

  PUBMED: 22885924; PMC: 3433394; DOI: 10.1038/ng.2385

The following data can be downloaded (updated April 2019):

• MAGIC_Metabochip_Public_data_release_25Jan.zip

Glucose and insulin results accounting for BMI

Glucose results accounting for BMI are from an analysis of 29 studies in up to 58,074 non-diabetic participants and the insulin results accounting for BMI are from an analysis of 26 studies in up to 51,750 non-diabetic participants. Details of the analyses accounting for BMI are within the zip file containing the data.

Please cite:

• A genome-wide approach accounting for body mass index identifies genetic variants influencing fasting glycemic traits and insulin resistance.

  Manning AK et al.

  Nature genetics 2012;44;6;659-69

  PUBMED: 22581228; PMC: 3613127; DOI: 10.1038/ng.2274

The following data can be downloaded:

• MAGIC_Manning_et_al_FastingGlucose_MainEffect.txt.gz
• MAGIC_Manning_et_al_FastingGlucose_Interaction.txt.gz
• MAGIC_Manning_et_al_lnFastingInsulin_MainEffect.txt.gz
• MAGIC_Manning_et_al_lnFastingInsulin_Interaction.txt.gz
• MAGIC_Manning_et_al_HOMAB_MainEffect.txt.gz
• MAGIC_Manning_et_al_HOMAIR_MainEffect.txt.gz
• MAGIC_Manning_et_al_README.pdf

Fasting proinsulin datasets

The fasting proinsulin datasets were generated by a meta-analysis of GWAS data in 10,701 non-diabetic adults of European ancestry. Fasting proinsulin values are adjusted for fasting insulin, age, sex and study-specific covariates.

Please cite:

• Genome-wide association identifies nine common variants associated with fasting proinsulin levels and provides new insights into the pathophysiology of type 2 diabetes.

  Strawbridge RJ et al.

  Diabetes 2011;60;10;2624-34

  PUBMED: 21873549; PMC: 3178302; DOI: 10.2337/db11-0415

The following data can be downloaded:

• MAGIC_ln_fastingProinsulin.txt.gz

HbA₁c association results

The HbA₁c association results were available in up to 46,368 non-diabetic adults of European descent from 23 GWAS and combined using inverse-variance meta-analysis. HbA₁c trait values are untransformed and adjusted for age, sex and study-specific covariates.

Please cite:

• Common variants at 10 genomic loci influence hemoglobin A₁(C) levels via glycemic and nonglycemic pathways.

  Soranzo N et al.

  Diabetes 2010;59;12;3229-39

  PUBMED: 20858683; PMC: 2992787; DOI: 10.2337/db10-0502

The following data can be downloaded:

• MAGIC_HbA1C.txt.gz

2hr glucose datasets

The 2hr glucose datasets were generated by a meta-analysis of nine GWAS in 15,234 non-diabetic individuals and a follow-up of 29 independent loci in 6,958-30,620 individuals. Trait values for 2hr glucose are untramsformed and are adjusted for BMI in addition to age, sex and study-specific covariates.

Please cite:

• Genetic variation in GIPR influences the glucose and insulin responses to an oral glucose challenge.

  Saxena R et al.

  Nature genetics 2010;42;2;142-8

  PUBMED: 20081857; PMC: 2922003; DOI: 10.1038/ng.521

The following data can be downloaded:

• MAGIC_2hrGlucose_AdjustedForBMI.txt

Fasting insulin and fasting glucose datasets

The fasting insulin and fasting glucose datasets were generated by performing a meta-analysis of up to 21 genome-wide association studies (GWAS) informative for fasting glucose, fasting insulin and indices of β-cell function (HOMA-B) and insulin resistance (HOMA-IR) in up to 46,186 non-diabetic participants. Fasting glucose trait values are not transformed. Trait values for Fasting insulin, HOMA-IR, HOMA-B and fasting proinsulin have been naturally log transformed. All datasets are adjusted for age, sex and study-specific covariates.

Please cite:

• New genetic loci implicated in fasting glucose homeostasis and their impact on type 2 diabetes risk.

  Dupuis J et al.

  Nature genetics 2010;42;2;105-16

  PUBMED: 20081858; PMC: 3018764; DOI: 10.1038/ng.520

The following data can be downloaded:

• MAGIC_ln_FastingInsulin.txt
• MAGIC_FastingGlucose.txt
• MAGIC_ln_HOMA-B.txt
• MAGIC_ln_HOMA-IR.txt