A global research team maps how rare and common gene variants jointly drive obesity and metabolic disease, offering fresh clues for equitable, personalized prevention and treatment.
Study: Discovery of obesity genes through cross-ancestry analysis. Image Credit: jittawit21 / Shutterstock
In a recent study published in the journal Nature Communications, a group of researchers identified and validated genes associated with Body Mass Index (BMI) across multiple ancestries using rare protein-truncating variants (PTVs), and mapped their relationships with comorbidities, plasma proteins, and polygenic risk.
Global Obesity Trends and Genetic Gaps
One in eight adults worldwide now lives with obesity, and small changes on the scale can ripple into diabetes, heart failure, and joint disease at home and in clinics. Why some people gain weight more easily remains partly genetic, yet most discoveries have come from European-only cohorts that do not reflect global diversity. BMI simplifies weight into a number, but the biology spans brain appetite circuits, adipose tissue, hormones, and the environment. Understanding rare, high-impact variants alongside the common polygenic burden can refine prevention and treatment strategies. Further research should move beyond single-ancestry designs to ensure genetic findings translate equitably.
Cross-Ancestry Study Design and Data Sources
The investigators analyzed 839,110 adults from the United Kingdom Biobank (UKB) and All of Us (AoU) spanning six continental ancestries. They performed ancestry inference using principal components and random forests, then ran gene-based rare-variant association tests with REGENIE v3.317, collapsing PTVs and deleterious missense variants into masks per gene. BMI was inverse-normal transformed within ancestry and sex strata. Covariates included age, age-squared, sex, age by sex, exome release batch (rather than genotyping batch), and the first ten genetic principal components. Results were meta-analyzed with inverse-variance-weighted fixed effects to create European, non-European, and all-ancestry summaries, with heterogeneity assessed by Cochran’s Q.
Statistical Analyses and Sensitivity Testing
Sensitivity checks included leave-one-variant-out analyses and conditioning on nearby common-variant signals. Phenome-wide association study (PheWAS) models tested clinical diagnoses. Structural equation modeling (SEM) was used to evaluate whether BMI mediated gene-comorbidity paths. Polygenic scores (PGS) for BMI quantified common-variant burden and its interaction with rare variants. Plasma proteomics in approximately 50,000 UKB participants linked PTV-carrier status to circulating proteins. The functional context was drawn from the Genotype-Tissue Expression project (GTEx), the International Mouse Phenotyping Consortium (IMPC), and the Common Metabolic Diseases Knowledge Portal (CMDKP).
Discovery of Novel BMI-Associated Genes
Thirteen genes reached exome-wide significance for association with BMI, including established loci such as melanocortin 4 receptor (MC4R), bassoon presynaptic cytomatrix protein (BSN), and proprotein convertase subtilisin/kexin type 1 (PCSK1), and five previously unreported associations: replication timing regulatory factor 1 (RIF1), YLP motif containing 1 (YLPM1), GRB10 interacting GYF protein 1 (GIGYF1), solute carrier family 5 member 3 (SLC5A3), and glutamate metabotropic receptor 7 (GRM7). After controlling for genomic inflation, associations for SLC5A3 and GRM7 were at suggestive significance, with effect sizes consistent with those observed in the primary analysis. Effect sizes for several of these rare protein-truncating variant burdens were comparable to those of canonical obesity genes, with YLPM1 and RIF1 showing magnitudes similar to those of MC4R and BSN. YLPM1, MC4R, and SAFB-like transcription modulator (SLTM) showed consistent effects in European and non-European groups, indicating broad generalizability. No gene reached exome-wide significance within any single non-European ancestry stratum, underscoring limited power for ancestry-specific discovery. In contrast, GRM7 showed significant heterogeneity by ancestry, and APBA1 displayed a European-biased effect, highlighting how ancestry composition can shape discovery and therapeutic targeting.
Gene-Carried Risks for Obesity and Comorbidities
Clinically, carriers of PTVs in YLPM1, RIF1, GIGYF1, and GRM7 had higher odds of being in the obesity and severe obesity categories, whereas SLC5A3 did not show enrichment for these categories. Beyond weight, cardiometabolic comorbidity patterns emerged. For example, BSN and GIGYF1 carriers exhibited elevated risk for type 2 diabetes, with BSN carriers also showing higher risk of hypertension and heart failure. SEM indicated that the risk of type 2 diabetes in BSN, GIGYF1, and SLTM carriers reflected both a direct pathway from gene to disease and an indirect pathway mediated by BMI. Conversely, the risk of gastroesophageal reflux disease (GERD) in SLC5A3 carriers appeared to be independent of BMI, suggesting that SLC5A3 increases GERD risk via a BMI-independent path and distinct mechanisms for the symptom clusters that patients experience.
Pleiotropy and Functional Validation Across Systems
A PheWAS expanded the view of pleiotropy. YLPM1 carrier status is associated with altered mental status and cholelithiasis, while GIGYF1 is associated with hypothyroidism and chronic renal failure alongside type 2 diabetes. Expression and model-organism data placed most of the implicated genes in brain regions and adipose tissue, consistent with central nervous system regulation of appetite, energy balance, and downstream adiposity. Heterozygous knockout of Ylpm1 in mice increased body fat mass and fasting glucose, supporting a causal role.
Combined Effects of Rare Variants and Polygenic Risk
PGS analyses showed additive effects between common-variant burden and rare high-impact variants. Among carriers, obesity prevalence rose steadily from the lowest to highest PGS quintiles in both UKB and AoU, and mean BMI was substantially higher in carriers with high polygenic load. This pattern suggests that everyday risk accrues from many small variants even when a single rare variant is present.
Plasma Protein Pathways Linked to Obesity Genes
Plasma proteomics identified rare variants associated with circulating proteins linked to metabolic health. SLTM carrier status is associated with higher leukocyte cell-derived chemotaxin 2 (LECT2), a hepatokine that tracked positively with BMI in the broader cohort. GIGYF1 carriers had lower odontogenic ameloblast-associated protein (ODAM) and neurocan (NCAN), and bridge-like lipid transfer protein family member 1 (BLTP1) carriers had higher cluster of differentiation 164 (CD164) and tumor necrosis factor superfamily member 12 (TNFSF12), with all of these protein levels also related to BMI. These relationships nominate downstream protein pathways that could be monitored or targeted.
Implications for Global Obesity Research
Across six ancestries and two large biobanks, the study identified five previously unreported obesity-associated genes and clarified which signals are generalizable across populations. The findings connect rare PTVs, common polygenic background, and plasma proteins to real-world outcomes such as type 2 diabetes, hypertension, heart failure, and gastroesophageal reflux disease. For families and clinicians, this means risk is not destiny but a layered combination of high-impact and polygenic influences. For researchers and developers, cross-ancestry sampling is crucial to avoid biased targets and promote equity.