Periodic Table of Food Initiative
A global, multi-organizational effort to map out the biomolecular diversity of our food
PTFI is what happens when a global network of stakeholders come together to better understand the “nutritional dark matter” of food. Indeed, the Periodic Table of Food Initiative aims to catalog the biomolecular composition of the world’s food supply.
Dayspring’s president, Chi-Ming Chien, is a co-founder of Verso Biosciences, Dayspring’s joint venture for life science software. Joining with sector leaders and domain experts from around the globe, we leverage the Dayspring team’s deep expertise in user-centered software, data processing, and performant scaling in order to lead the data workstream within the initiative.
The PTFI: where foodomics meets scale
The goal? To catalog a deep knowledge of food composition in a publicly available database in order to tackle bigger questions surrounding human and planetary health.
Organizational leaders and domain experts must endeavor together within different workstreams in order to provide both quantitative and qualitative support to the project. After all, researchers, nutritionists, and farmers all hold a different piece of the puzzle when it comes to the global food ecosystem. How does climate change impact the quality of food? In what ways can we empower data-driven decisions to aid the production of food that is nutrient-dense, desirable, and affordable? Why do sustainable farming practices matter when it comes to improving food security and nutritional health?
To answer these bigger questions, standardized tooling and analytics combine with a system for centralized data processing.
Advanced science and software together
To start, labs need to upload food sample data generated using the latest mass spectrometry technology. To facilitate and expedite the process, Dayspring designed and built the “Verso Lab Portal,” an easy-to-use web application interface for uploading data and receiving results.
The focus of much of food composition measurement has been on the common macronutrients and other well-known minerals and vitamins. These, however, form only a fraction of what’s actually in food. PTFI is working to characterize everything down to specialized metabolites, small molecules like flavonoids and terpenoids, that may have significant health effects. By merging analytical data about the compounds’ polarity, mass, shape, and fragmentation patterns, we can record and begin identifying the “dark matter” composition of foods we eat every day.
But here is where cloud computing and data pipeline processing comes into play. The computing power required for processing all of this data is massive, and well beyond what each individual lab might typically have access to. A traditional lab might process each sample in series and take days or even weeks to process a batch. By leveraging cloud services and stratified processing in parallel, we’re able to process hundreds or thousands of samples across thousands of computing cores—all in a pipeline that does all the heavy lifting and coordinating for these data-intensive computations.
Furthermore, mass spectrometry processes and liquid chromatography measurements can vary from lab to lab across the globe. The pipeline takes care of standardization: merging metadata with the analytical data and allowing data generated in labs around the globe to be aggregated and compared with one another in a centralized food compound database.
Finally, it’s the data visualization that truly allows researchers to glean hands-on insights with the data. Using a branded version of MarkerLab, a custom web application we developed with Verso for organizing and visualizing research data, these datasets get transformed into customizable, interactive reports. This is where we’re able to extract benchmarks and grouped data comparisons on a whole suite of different compound measurements. Insights and standardized metric comparison for thousands of these different compounds enables us to understand their roles in broader nutrition, health, and ecosystem functioning and powers scientific discovery.
“…most compounds in the plants, meat, fish, and dairy products that we eat have never even been named and represent the “dark matter of food…””
“This platform allows us to analyze the complexity of foods in a way that was simply not possible before…”
Chi-Ming Chien in an article for Science.org / American Association for the Advancement of Science
“[PTFI data via MarkerLab… will allow researchers] to compare foods and compounds of interest across the entire PTFI database to spark the next scientific discoveries around food, nutrition, agriculture and health.”
Chi-Ming Chien in an article for Forbes