Communications of the ACM
The Food Pyramid Is Out
An overview of the complexity of the human microbiome; circle size indicates approximate relative abundance of microbes known to associate with humans.
Credit: Human Microbiome Consortium
We are not alone in our own bodies; we co-exist with 10 times as many bacteria as human cells, and by analyzing and manipulating the bacteria, they could help guide us to better health.
The cost of microbial genome sequencing plummeted a decade ago, thanks in part to the development of bioinformatics/computational tools to improve sequencing pipelines and to improve downstream sequence analysis. Add to that the success of the Human Microbiome Project (a U.S. National Institutes of Health initiative launched in 2008 to better understand how health and disease are affected by changes in the human microbiome) and a call from the White House early this year to support microbiome science, and top it off with $1.5 billion in investment by venture capitalists in biotechnology in 2015 (second only to software investment), and the result appears to be great opportunity to commercialize personalized (or "bespoke") diets based on microbiome analyses.
That has not happened yet. There has been a profusion of academic studies investigating the human gut microbiome, which used to be referred to as "gut flora" ("microbiome" commonly means the environment in which microorganisms and their genomes live, and the term "microbiota" is often used synonymously). It is also true some ventures appear to be on the cusp of commercializing products or tools that can map a person’s microbiome in order to offer a personalized diet to maximize health. Yet companies are reticent to trumpet any successes in the field.
One organization that appears to be making significant progress is the Personalized Nutrition Project (PNP) at Israel’s Weizmann Institute of Science. William Bonificio, co-founder of the American Microbiome Institute, a Somerville, MA-based non-profit dedicated to advancing microbiome science and education, said PNP is "better than anyone else when it comes to personalizing the microbiome at the consumer level." George Cigale, an executive in the microbiome field who met with Eran Elinav, one of PNP’s principals, agreed PNP is "farther ahead than others in the field; close to the best."
PNP has found people respond to the same foods differently, based on the makeup of the bacteria in their gut microbiome; for some people, eating ice cream was healthier than eating sushi. By monitoring glucose levels in 800 study participants and measuring the responses to almost 47,000 meals, as published in Cell, the project found, "high variability in the response to identical meals, suggesting that universal dietary recommendations may have limited utility." The study "devised a machine-learning algorithm that integrates blood parameters, dietary habits, anthropometrics, physical activity, and gut microbiota…and showed that it accurately predicts personalized postprandial glycemic response to real-life meals." The researchers validated their predictions "in an independent 100-person cohort. Finally, a blinded randomized controlled dietary intervention based on this algorithm resulted in significantly lower postprandial responses and consistent alterations to gut microbiota configuration." The results suggest, "…personalized diets may successfully modify elevated postprandial blood glucose and its metabolic consequences."
PNP appears to be on the verge of going commercial. Its website states, "the research technology was licensed to DayTwo," although reaching out to DayTwo resulted in an email response that stated, "Our product is currently under development, and its commercial launch is planned to take place in the second half of 2016. We are making all the necessary efforts to complete the development stage in order to meet the extremely high demand." However, it is likely news will come out sooner because Lihi Segal, DayTwo’s CEO, will be speaking at the Translational Microbiome Conference on "Personalized Nutrition Using Gut Microbiome and Clinical Data" in Boston on April 20. The synopsis states, "DayTwo uses and extends the algorithm to provide personalized nutrition and actionable insights that allow people to maintain normal blood sugar levels." In addition, Eran Segal, the other principle scientist on the original PNP, will speak at the Personalized Nutrition Congress, also in Boston, in May.
Other organizations using computer technology that are on the cusp of bespoke diet tool launches include the work at Chalmers University of Technology in Sweden, and Bespoke Diet and Fitness (BD&F) in Canada.
Jens Nielsen, professor of systems biology and head of the research team at Chalmers, used an algorithm that identifies "in detail how some common intestinal bacteria interact during metabolism." Nielsen was loath to talk about the project, saying only in an email, "We are working on developing models that will eventually allow for quantitative analysis of diet effects. However, this is still early times."
BD&F, which collaborated with the U.K. firm DNAFit in 2015, commercialized a genetic test to create a personalized nutrition and conditioning program to "maximize your genetic potential," especially when it comes to maximizing fitness. The company’s web site says that, based on DNA swabs, it "tailors your unique genomic information (DNA) to create a personalized nutrition and conditioning program to maximize your genetic potential. Testing is performed in compliance with the highest international standards."
Time will tell if bespoke diets based on algorithms will take off. It is notable that, as Cigale observed, there are almost 67,000 dieticians and nutritionists in the U.S. and almost 13,000 gastroenterologists who advise their clients and patients on diet and nutrition. Both professions are growing, and a tool that would make it easier to customize personalized diet plans for clients or patients can only bring big wins for the sick, obese, or lethargic.
Tatjana Meerman is a freelance technology writer based in the Washington, D.C. area.
No entries found