21st century toolkit for optimizing population health through precision nutrition

Abstract

Scientific, technological, and economic progress over the last 100 years all but eradicated problems of widespread food shortage and nutrient deficiency in developed nations. But now society is faced with a new set of nutrition problems related to energy imbalance and metabolic disease, which require new kinds of solutions. Recent developments in the area of new analytical tools enable us to systematically study large quantities of detailed and multidimensional metabolic and health data, providing the opportunity to address current nutrition problems through an approach called Precision Nutrition. This approach integrates different kinds of "big data" to expand our understanding of the complexity and diversity of human metabolism in response to diet. With these tools, we can more fully elucidate each individual's unique phenotype, or the current state of health, as determined by the interactions among biology, environment, and behavior. The tools of precision nutrition include genomics, metabolomics, microbiomics, phenotyping, high-throughput analytical chemistry techniques, longitudinal tracking with body sensors, informatics, data science, and sophisticated educational and behavioral interventions. These tools are enabling the development of more personalized and predictive dietary guidance and interventions that have the potential to transform how the public makes food choices and greatly improve population health.

Keywords: Precision nutrition; big data; food science and technology; phenotyping.

Figure 1.

Overview of Precision Nutrition research and intervention.Data sets arising from analyses of human clinical samples, genetics, electronic medical records, and monitoring of dietary intake, physical activity and behavior are used to develop personal nutrition predictive algorithms. Modeling of observations from thousands of people is used to predict expected health outcomes for new subjects and determine what dietary changes should be recommended based on each person’s individual data and phenotype.