Multimodal phenotyping reveals plant-food-related dietary information in serum Raman spectra of community-dwelling Japanese adults
This study demonstrates that in community-dwelling Japanese adults, habitual plant-food-related dietary information is not captured by a single biological state but is instead represented as traceable dimensions distributed across complementary phenotypic layers, with serum Raman spectra providing unique predictive insights into specific dietary variables like beta-carotene that complement clinical and metabolomic data.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
For decades, doctors and scientists have understood that what we eat shapes our long-term health. We know that a diet rich in vegetables, fruits, and whole grains lowers the risk of heart disease and diabetes, while diets heavy in processed meats and sugary drinks do the opposite. To measure this, researchers often use scoring systems that tally up how many servings of healthy foods a person eats versus unhealthy ones. These scores give a single number representing a person's overall diet quality. However, a single number cannot tell the whole story. Two people might have the exact same score but achieved it through completely different combinations of foods. Furthermore, even when people eat the same foods, their bodies react in unique ways; one person's blood sugar might spike after a meal while another's remains steady. The big question for modern science is how to see these subtle, individual differences in the body's biology. We know diet leaves a mark on our blood, but we are still learning exactly where that mark appears and how to read it.
A team of researchers in Japan recently tackled this puzzle by looking at the blood of nearly four hundred community-dwelling adults. They wanted to see if they could find specific traces of what these people habitually ate hidden inside their blood samples. The team collected detailed records of what the participants ate over time and then took fresh blood samples to run three different types of tests. First, they performed standard blood tests that doctors use every day to check for things like cholesterol and sugar levels. Second, they analyzed the serum metabolome, which is a detailed chemical map of the tiny molecules floating in the blood that result from digestion and metabolism. Third, they used a technique called Raman spectroscopy. This method shines a specific type of light on the blood serum and measures how the light bounces off the molecules. It acts like a molecular fingerprint, revealing the chemical composition of the sample without needing to add any dyes or labels.
The researchers started by grouping the participants based on their diet quality scores. They found that people with higher scores, indicating healthier eating habits, generally had a better health history over the previous several years. These individuals had fewer hospitalizations and better numbers on standard health checks like blood pressure and kidney function. However, when the team looked at the complex biological data from the blood tests and the chemical maps, they did not see the groups separating into distinct, clear clusters. In other words, a healthy diet did not create a single, obvious "healthy" biological state that looked completely different from an "unhealthy" one. The bodies of people with good diets and those with poorer diets looked surprisingly similar when viewed as a whole, suggesting that diet does not simply switch the body into one mode or another.
The story changed, however, when the researchers focused on specific foods rather than the overall score. They asked a more precise question: could they predict how much of a specific food, like coffee or a certain vegetable, a person ate just by looking at their blood data? They found that the standard blood tests and the chemical maps could predict some of this, but the Raman spectroscopy added a crucial piece of the puzzle. For many specific dietary variables, including coffee and certain plant-based nutrients, the light-scattering patterns in the blood provided extra information that the other tests missed. This was particularly clear for beta-carotene, a nutrient found in orange and yellow vegetables. The Raman patterns associated with beta-carotene could be traced directly to specific vibrations in the molecules known to be characteristic of carotenoids. This confirmed that the light was indeed picking up the physical presence of these plant-based compounds in the blood.
What made this discovery even more significant was that the signal for beta-carotene was not an isolated event. The same patterns in the light that pointed to beta-carotene also aligned with a broader group of healthy, plant-based foods. When the researchers mapped out the direction of these signals, they found that vegetables, fruits, legumes, and the overall healthy diet score all pointed in the same direction. Conversely, signals for less healthy items like processed meats and sugary drinks pointed in the opposite direction. This suggests that the body does not just store a single "diet score" in the blood. Instead, the influence of food is distributed across different layers of biology. Some aspects of diet are visible in standard blood tests, some in the chemical maps, and some are uniquely captured by the way light interacts with the blood.
The study concludes that our habitual diet is not represented in the body as a single, switch-like state. Instead, it is woven into the biology as a series of traceable dimensions that are spread across different types of measurements. While a simple diet score might not separate people into two distinct biological groups, the specific details of what we eat are still there, waiting to be found if we look with the right tools. By combining standard medical tests with advanced light-based scanning, scientists can now see a more complete picture of how food shapes our biology, moving beyond simple scores to understand the specific, traceable footprints of the meals we eat.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.