Modelling BMI trajectories in children for genetic association studies

Area of Science:

• Genetics
• Pediatrics
• Obesity Research

Background:

• Childhood growth patterns and genetic predispositions are key to understanding adult obesity.
• Identifying genetic effects on longitudinal growth requires robust statistical models.

Purpose of the Study:

• To define statistical models for detecting genetic effects on childhood growth.
• To investigate associations between common genetic variants and BMI trajectories in children.

Main Methods:

• Genotyped 1,506 children from the Raine Study at 17 obesity-associated genetic loci.
• Calculated an obesity-risk-allele-score for each individual.
• Compared four statistical models (linear mixed effects, skew-t errors, semi-parametric, non-linear mixed effects) for BMI growth profile analysis.

Main Results:

• The semi-parametric linear mixed model proved most efficient for detecting genetic effects on childhood growth.
• Three loci were significantly associated with BMI in females, and four in males.
• The obesity-risk-allele score correlated with increased average BMI and growth rate in both sexes.

Conclusions:

• Appropriate statistical models reveal genetic influences on childhood growth related to adult obesity genes.
• Significant sex-based differences were observed in genetic associations with BMI trajectories.
• This research offers insights into the biology of childhood growth and identifies early-life genetic markers for BMI changes.