Obesity, a condition influenced by genetic, environmental, and behavioral factors, poses significant health risks. This study leverages the GenObIA dataset, collected from January 2018 to June 2022, which includes extensive sociodemographic, environmental, behavioral, and genetic data, to model the risk of overweight and obesity using genetic programming techniques. Our aim is to develop predictive models to identify individuals at risk of these conditions, thereby facilitating early intervention and prevention strategies. Through structured grammatical evolution and symbolic regression by genetic programming, we generate interpretable machine learning models that reveal the influence of genetic variants, particularly single nucleotide polymorphisms, on obesity. Our findings underscore the importance of age, metabolic syndrome, and lifestyle factors such as alcohol consumption in predicting obesity. The models indicate potential, but further refinement in data preprocessing and model training is necessary to improve prediction reliability. This research contributes to a deeper understanding of obesity and supports the development of targeted public health interventions.