Evidence-based precision nutrition improves clinical outcomes by analyzing human and microbial molecular data with artificial intelligence

Abstract

Many studies have shown that foods and nutritional ingredients play an important role in healthy human homeostasis, either directly or via the microbiome. We have developed an objective, integrated, and automated approach to personalized food and supplement recommendations that is powered by artificial intelligence and individualized molecular data from the gut microbiome, the human host, and their interactions. The process starts with a clinically validated transcriptomic analysis of a persons stool (and some cases also blood) sample, from which the following molecular data are computed using clinically validated bioinformatics: 1. Quantitative strain-level taxonomic classification of all microorganisms, 2. Gene expression level for microbial genes, 3. Gene expression level for all human genes. These molecular data are converted into personalized nutritional recommendations (foods and supplements) using algorithms derived from clinical research studies and domain knowledge. We describe an application of our precision nutrition technology platform to human populations with irritable bowel syndrome (IBS), depression, anxiety, and type 2 diabetes (T2D). In these pilot interventional studies, our precision nutrition program achieved significant improvements in clinical outcomes of IBS (39% for severe IBS), depression (31% for severe depression), anxiety (31% for severe anxiety), and the risk score for T2D (>30% reduction relative to the control arm). These data support the integration of data-driven precision nutrition into the standard of care.