Methodological aspects of the assessment of gene–nutrient interactions at the population level

Nutritional-related diseases are the result of complex interactions between genes and diet. The understanding of these interactions will provide the rationale for dietary interventions based on the individual's genetic constitution. However, the approach to this kind of study is not easy, the complexity of the interactions increasing exponentially the dimensionality of the problem. The aim of this review is to analyze the major problems that arise in approaching complex interactions at the population level. Furthermore, several statistical tools available for this type of analysis are discussed. In conclusion, although analytic techniques able to reduce the dimensionality of the problem are suggested, sample size requirement seems to remain an inescapable challenge for the researcher. A synergy between traditional and nontraditional statistical approaches could be useful.