Why two people following identical diets can have vastly different outcomes has long puzzled nutrition scientists. Emerging evidence points to a surprisingly underexplored factor: the genetics of how we taste and smell food. If chemosensory perception is partly heritable and shapes what people eat, it may also explain why behavioral nutrition interventions succeed for some and fail for others — an insight with real implications for precision dietary medicine.
This review, published in the Proceedings of the Nutrition Society, synthesizes evidence linking genetic variation in chemosensory pathways to dietary behavior and nutritional outcomes. Twin and family studies indicate moderate-to-high heritability for bitter taste perception specifically, while heritability estimates for sweetness, sourness, saltiness, fat-related taste traits, and olfactory sensitivity are more modest and phenotype-dependent. Genome-wide association studies (GWAS) have identified loci in taste receptor genes — particularly those encoding TAS2R bitter taste receptors — and olfactory receptor genes, associating them with food liking and dietary intake patterns. The review also examines how Mendelian randomization, which leverages genetic variants as natural experiments, has been applied to test whether sensory phenotypes causally influence eating behavior rather than merely correlating with it.
The field sits at an intellectually productive crossroads between sensory science, genomics, and nutritional epidemiology, yet several important caveats temper enthusiasm. The GWAS evidence remains heavily skewed toward bitter taste and populations of European ancestry, limiting generalizability. Chemosensory phenotyping is rarely conducted at the scale of tens or hundreds of thousands of participants needed for robust genomic discovery. Mendelian randomization offers causal leverage but depends on instruments that may not fully capture the complexity of polygenic sensory traits. The practical translation to personalized nutrition — matching dietary recommendations to an individual's sensory genotype — remains aspirational rather than clinical-ready. Still, this review is a meaningful consolidation of a field gaining momentum, and the framework it proposes for integrating genomic sensory data into precision nutrition trials represents a genuinely useful methodological roadmap for future research.
