magic.scores: Genetic risk scores for glucose/insulin traits.


Risk scores parameterised using GWAS meta-analysis results published by the MAGIC consortium (Dupuis et al. Nature Genetics 2010) for glucose and insulin traits.




A data frame suitable for use with other functions in this package, see gtx.params.

The scores are called FG2010 and FI2010, for fasting plasma glucose (FG, mmol/l) and fasting insulin (FI, pmol/l), and using the 16 and 2 genome wide significant loci from Table 1 of Dupuis et al. (2010) respectively. Effect size coefficients are taken from Table 2 of Dupuis et al. (2010) and hence use SNPs different from those reported in Table 1 at the SLC30A8 and TCF7L2 loci.

In addition, for all SNPs in both risk scores, there are effect size estimates (beta) and standard errors (se) for FG, HOMA-B index of beta cell function, FI, HOMA-IR index of insulin resistance, glycated hemoglobin (HbA1c, percent), glucose 2 hours after oral glucose tolerance test (2hG, mmol/l), insulin 2 hours after oral glucose tolerance test (2hI, pmol/l), type 2 diabetes risk (T2D, ln odds, converted from odds ratio), body mass index (BMI, kg/m2), diastolic blood pressure (DBP, mmHg), systolic blood pressure (SBP, mmHg), hypertension risk (HTN, ln odds), and serum lipid concentrations (HDL, LDL, TC, TG), all taken from Table 2 and Table 3 of Dupuis et al. (2010).


For the publication by Dupuis et al. from which these data were extracted see


with(subset(magic.scores, score = "FG2010"),
  grs.plot(beta_FG, beta_TG, se_TG, text = locus))

Questions? Problems? Suggestions? or email at

All documentation is copyright its authors; we didn't write any of that.