A large genetic study has identified 624 independent genetic signals associated with adiposity during childhood, revealing that the biology influencing body size early in life is not simply a smaller-scale version of adult obesity genetics. About one-third of the identified variants showed no concordant association with adult body mass index, suggesting that some genetic effects are particularly important during childhood.
The study, published in Nature Genetics, combined longitudinal childhood measurements, recalled childhood body size, puberty-related genetic data, whole-genome sequencing and single-nucleus RNA sequencing. The findings point to age-specific roles for endocrine and neuropeptide pathways involved in energy balance, including leptin-melanocortin and incretin signalling.
A genetic map across childhood
The researchers first conducted age-stratified genome-wide association analyses using data from the Norwegian Mother, Father and Child Cohort Study. Body mass index was assessed at 11 time points from six weeks to eight years of age, with up to 62,276 children contributing measured longitudinal data.
To increase statistical power, the team then used genomic structural equation modelling to combine childhood BMI at age eight, recalled body size at age 10 from UK Biobank, and genetic data related to age at menarche. This produced an effective sample size of 599,924 for the childhood adiposity common factor.
Across the age-specific and combined analyses, the researchers identified 624 independent genetic signals. Only 44, or 7%, had previously been discovered in the two largest genetic studies based on measured childhood BMI.
Many genetic effects changed with age
The researchers found substantial variation in how genetic effects appeared across childhood and adulthood. Roughly one-third of the 624 signals did not show a concordant association with adult BMI, highlighting genetic influences that may be more specific to early life.
Several signals connected to the leptin-melanocortin pathway, including BSX, GNAS, LEPR and PCSK1, showed childhood-specific effects. Signals involving the incretin-related genes GIPR and GLP1R also displayed effects that differed between childhood and adulthood.
A polygenic score based on 526 signals from the childhood adiposity common factor explained up to 9.2% of the variation in childhood BMI, peaking at age nine. It also explained up to 6.1% of variation in childhood body fat percentage at the same age. These childhood-focused scores performed better during childhood than comparable scores based on adult BMI genetics.
Rare variants strengthened the childhood-specific picture
The study extended beyond common genetic variants. Whole-genome sequencing data from 479,615 individuals identified rare protein-coding variation in ADCY3, CALCR, MC4R, MRAP2, POMC and MYH13 that showed stronger associations with adiposity in childhood than in adulthood.
This matters because several of these genes are already implicated in biological systems controlling appetite, energy balance and body weight. The age-specific results indicate that studying children can reveal genetic effects that may become weaker or less visible later in life.
Brain cell data pointed to childhood-specific pathways
The researchers also integrated single-nucleus RNA sequencing data from the human hypothalamus. They identified childhood-specific adiposity-regulating cell populations in the arcuate nucleus and mammillary bodies, two brain regions involved in metabolic and behavioural regulation.
Genes contributing to the classification of these childhood-specific cell populations included components of the leptin-melanocortin pathway such as LEPR, POMC, BDNF, MC4R and PCSK1, as well as genes linked to incretin and gut-peptide signalling including GLP1R, GIPR and GRP.
What the findings mean
The results strengthen the case for treating childhood as a distinct biological period in obesity research. Genetic studies based predominantly on adult BMI can identify many important pathways, but they may miss variants whose effects are strongest during infancy or childhood.
The study does not imply that genes determine whether an individual child will develop obesity. Body weight reflects interactions among genetic susceptibility, development, behaviour, environment and broader social conditions. Instead, the findings help clarify biological pathways that influence how adiposity develops across different stages of life.
The age-specific signals may eventually help researchers understand why particular biological pathways are more influential at certain developmental stages and whether prevention or treatment approaches could be better aligned with those stages. Clinical applications, however, require considerably more evidence.
Important limitations
The childhood adiposity datasets were drawn from cohorts of northern European ancestry. The researchers therefore emphasised the need for more genetically diverse populations before the findings can be assumed to generalise broadly.
BMI was also used as a marker of adiposity during infancy and childhood. Although BMI is useful in large population studies and is supported by comparisons with more direct body-composition measures, it does not distinguish fat mass from lean mass with the precision of dedicated body-composition techniques.
Some of the combined analysis also relied on adults recalling their body size at age 10. In addition, the cell-type analyses cannot by themselves establish that changes in gene expression cause the observed age-specific genetic effects. Further work using childhood tissue samples, more diverse cohorts and age-specific experimental models will be needed.
Source Information
Study: Genome-wide mapping of common and rare variant effects on adiposity across childhood
Journal: Nature Genetics
Publication year: 2026
Study design: Age-stratified genome-wide association analyses, genomic structural equation modelling, polygenic scoring, whole-genome sequencing and single-nucleus RNA sequencing integration
Key samples: Up to 62,276 children with objectively measured longitudinal adiposity-related traits; effective sample size of 599,924 in the combined childhood adiposity analysis; whole-genome sequencing in 479,615 individuals
Primary finding: Researchers identified 624 common genetic signals associated with childhood adiposity, with approximately one-third showing no concordant association with adult BMI. Childhood-focused genetic scores explained up to 9.2% of childhood BMI variation.
Source: Nature Genetics. DOI: 10.1038/s41588-026-02772-y
