Login
A physicist and his team have developed a tool to help consumers determine how ultraprocessed a food is — research that could eventually help scientists understand how specific foods might prevent or treat disease.
Albert-László Barabási, a Hungarian-American physicist, professor and pioneer in network science, told the “Real Organic Podcast” that his lab has identified about 135,000 molecules naturally present in foods but not included in the standard nutritional picture.
Barabási calls these compounds “nutritional dark matter.” He believes understanding them could eventually transform the idea of “food as medicine” by revealing how individual foods and their chemical components affect the body.
Poor nutrition is a major contributor to chronic disease, including cardiovascular disease, Type 2 diabetes and cancer.
Yet nutrition science has focused largely on a relatively small number of nutrients, while thousands of other compounds in food remain poorly understood.
For consumers, one of his lab’s most immediately useful tools is already available.
‘We were shocked at how many people are using it’
Barabási’s research team developed GroceryDB, an artificial-intelligence (AI)-based tool that classifies foods according to their degree of processing.
The system analyzes a food’s nutritional components and determines its degree of processing with about 98% accuracy, Barabási said.
Researchers created GroceryDB as a research tool, but shoppers quickly began using it to compare foods.
The tool allows consumers to look up products sold at Whole Foods, Walmart, Target and other retailers to see how highly processed they are. It also compares products within the same category, allowing shoppers to find less-processed alternatives.
“We really put it out not as a consumer product,” Barabási said. “We were shocked at how many people are using it as a guide.”
The unexpected consumer interest points to a much larger question Barabási’s lab is pursuing: What is actually in the food people eat, and what does it do inside the body?
‘The biggest gorilla in the room at the cellular level is the food’
Barabási came to food research through his work studying networks of interactions inside cells and how disruptions in those networks can contribute to disease.
Genomic information can explain some disease risk, but it accounts for only about 10% of disease occurrence, he said.
“The question is, ‘Where is the rest?’” he asked.
Barabási said the missing piece is largely the environment, which includes factors such as sleep, stress and exercise.
“But the biggest gorilla in the room at the cellular level is the food,” because it delivers a huge number of molecules directly into the body, he said.
“That was the journey that eventually led me to start thinking about what food is made of,” Barabási said.
Nutrition science targets the relatively small number of nutritional components — about 150 — that humans need for energy and survival, including sugars, fats, vitamins and other essential nutrients.
But Barabási said he was surprised to discover that many chemicals known to be present in food are not included in conventional nutritional databases.
Barabási Lab, headquartered at Northeastern University in Boston, initially identified about 20,000 such compounds. It now has evidence of about 135,000 molecules found in one or more foods that are not considered nutritional components.
These are not human-made substances added to ultraprocessed foods, according to Barabási. These are chemicals that exist naturally in whole foods.
‘Nutritional dark matter’ could improve human health
Polyphenols make up one major group of these overlooked compounds. Found in plants, they include flavonoids that provide the color, taste and flavor of many fruits and vegetables.
Unlike conventional nutrients, these compounds do not primarily provide the body with energy. Instead, they can influence cellular activity by interacting with proteins and DNA, Barabási said.
“They act like drugs,” he said. “They turn on processes or turn off processes. … Their role is to really kind of modulate the cell’s activity.”
Scientists partly understand the role of several thousand of these compounds, Barabási said. But “for the vast majority, we do not.”
That is the “nutritional dark matter” his lab is trying to illuminate.
The researchers are using “the tools of network medicine to understand, for each of these molecules, what molecules it engages with in our cells, and then what it potentially does,” Barabási said.
The implications could extend well beyond simply learning more about the foods people eat.
Barabási pointed out that many existing medicines or supplements — such as vitamin C — originated in plants or foods. Some were derived directly from natural compounds, while others were created by modifying those compounds to make them more effective.
That raises a larger question: Are there other food-derived molecules that could improve human health?
“I think there are. I bet there are,” Barabási said. “We are searching for them.”
‘Can you tailor your diet in a way that is beneficial for a particular disease?’
That possibility fits directly into the growing “food as medicine” movement.
Dave Chapman, co-founder of the Real Organic Project and host of its podcast, noted that much of the current discussion focuses on avoiding foods that may be harmful, particularly ultraprocessed foods.
But there is another possibility, according to Barabási. Some foods may be especially beneficial, and some varieties of the same food may contain different amounts or combinations of compounds that matter for health.
Barabási said his lab wants to determine whether diets could be tailored to counter individual diseases and whether researchers could identify the molecules responsible for those positive effects.
The long-term possibility is to identify beneficial combinations of food-derived compounds and provide them as supplements or medicines.
“Can you tailor your diet in a way that is beneficial for a particular disease that you may already have?” he asked. “We are working towards that.”
The research could also lead to a new way of thinking about the food itself.
Instead of breeding or growing crops primarily for characteristics such as appearance and shelf life, producers could eventually select for chemical profiles associated with taste or health.
“If I can show that heart disease requires this combination of chemicals, if you could make a tomato that has that, and I’m at risk of heart disease, I want that,” Barabási said.
‘We are so far’ from knowing everything that’s in a tomato
The challenge is that scientists currently lack the tools to make many of those comparisons.
Barabási said a tomato may contain roughly 10,000 chemicals, but researchers have quantified only a small fraction of them. In other words, they may know that a compound exists without knowing how much of it is present.
That makes it difficult to determine why one tomato tastes better than another or whether growing conditions change its chemical profile in meaningful ways.
Taste, for example, depends heavily on volatile organic compounds and flavonoids that are not routinely measured in food.
That lack of data also makes it difficult to answer questions about differences between organically and conventionally grown foods.
Barabási said the two types of tomatoes would contain the basic chemical components that make a tomato a tomato. Any differences would involve the quantities of individual compounds — most of which researchers don’t measure.
“If we had really detailed methods to measure and quantify all the chemicals in a tomato,” then researchers could begin determining which chemical quantities are associated with organic production, better taste or other characteristics, he said.
“We are so far from that.”
‘When we process food, we actually change its composition’
Barabási’s work also highlights why the degree of food processing matters when evaluating what people actually consume.
An onion, for example, has one chemical composition when raw, he said. Boiling it changes some of its compounds, and frying it changes them further. By the time an onion is turned into an ultraprocessed product — say, a packaged onion ring — its chemical profile is dramatically different.
“It’s no longer recognizable as onion,” Barabási said. “When we process food, we actually change its composition.”
He distinguished ordinary food preparation from industrial ultraprocessing, noting that cooking methods used in a home kitchen generally do not fundamentally transform a food in the same way industrial ultraprocessing can.
“You need chemical factories … to achieve that,” he said.
This article was funded by critical thinkers like you.
The Defender is 100% reader-supported. No corporate sponsors. No paywalls. Our writers and editors rely on you to fund stories like this that mainstream media won’t write.
‘We are putting our efforts in the wrong place’
For consumers, GroceryDB offers a practical way to assess food processing now. But Barabási’s larger research focuses on questions that could take years to answer.
His lab wants to move beyond identifying what compounds are present in food to understanding how those compounds interact with human biology.
That could eventually allow scientists to identify foods or combinations of food compounds that are particularly beneficial for certain health conditions.
No dedicated government funding exists for systematically mapping the chemical composition of food, Barabási said.
Instead, research remains heavily focused on genomics, even though genes account for only a small part of disease risk.
“Humanity has not arrived to understand how important food is for health,” he said. “We are putting our efforts in the wrong place.”
Watch Barabási on the ‘Real Organic Podcast’ here:
Related articles in The Defender
- Let Food Be Thy Medicine? RFK Jr. Announces 50+ Medical Schools Will Add Nutrition Training
- Ultraprocessed Foods Make Up Over Half of Kids’ Diet, CDC Report Says
- RFK Jr.: MAHA Report a ‘Clarion Call’ to End the Chronic Disease Epidemic
- 3D-Printed Cookies Made From Plastic Bottles — Is This the Future of Food?
