Genetic studies have identified “favourable adiposity” variants - where the allele associated with higher adiposity is associated with lower risk of type 2 diabetes. We took a novel approach to find more of these alleles and find the underlying mechanisms.

We first performed a genome wide association study (GWAS) of body fat percentage using 451000 individuals from UK Biobank. Second, we used published genetic data in a multivariate test to find alleles associated with higher adiposity but a “favourable” metabolic phenotype: higher HDL-C, adiponectin, sex hormone binding globulin, but lower triglycerides, fasting insulin and alanine transaminase. Third, we used abdominal imaging data from 4 studies to define the adiposity phenotype in more detail.

We identified 620 variants associated with body fat percentage (p<5x10-8). Fourteen alleles had a “favourable adiposity” phenotype, including 7 previously known variants (in/near PPARG, LYPLAL1, GRB14, IRS1, PEPD, FAM13A and ANKRD55), 5 known to be associated with a relevant trait (TRIB1, KLF14/MKLN1, DNAH10, VEGFA/C6orf223 and AEBP2/PDE3A) and 2 novel variants (MAFF and CITED2). Carrying 10 additional "favourable adiposity" allele was associated with higher BMI (0.63 kg/m2, p = 4x10-45) but lower risk of type 2 diabetes (odds ratio: 0.60, p = 4x10-44, 14371 cases), lower risk of heart disease (0.82, p = 3x10-14, 37741 cases) and lower risk of hypertension (0.82, p = 1x10-33, 241691 cases). Using MRI/CT scans of abdominal fat from UK Biobank (N = 5000), NEO (2236), TÜF (900), and a published GWAS (10557), carrying more “favourable adiposity” alleles was associated with higher subcutaneous fat (p = 1x10-12) but lower liver fat (p = 1x10-7). These effects were similar when we removed the 7 known “favourable adiposity” variants.

“Favourable adiposity” alleles associated with higher BMI but lower risk of type 2 diabetes are characterized by higher subcutaneous but lower liver fat.


H. Yaghootkar: None. Y. Ji: None. A.M. Yiorkas: None. F. Frau: None. D. Mook-Kanamori: None. R. de Mutsert: None. J. Tyrrell: None. S.E. Jones: None. R. Beaumont: None. A.R. Wood: None. L. Thomas: None. K.V. Allebrandt: None. N. Stefan: Consultant; Self; Merck Sharp & Dohme Corp.. Speaker's Bureau; Self; Novo Nordisk A/S. Research Support; Self; AstraZeneca. Speaker's Bureau; Self; AstraZeneca, OmniaMed Ltd.. H. Staiger: None. E. Pearson: Speaker's Bureau; Self; Eli Lilly and Company. A.I. Blakemore: None. J.D. Bell: None. T.M. Frayling: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at