One of the most important parts of life science research is exploring the effects of specific dietary patterns on health. However, traditional methods of measuring diet (such as food diaries or food frequency questionnaires) not only require time and effort for research participa

One of the most important parts of life science research is exploring the effects of specific dietary patterns on health. However, traditional methods of measuring diet (such as food diaries or food frequency questionnaires) not only require time and effort for research participants to fill out, but more importantly, the accuracy of the data is difficult to guarantee and may directly affect the replicability of the research conclusions. sex.

For this reason, many scientists have turned to metabolomics. Metabolomics is an interdisciplinary discipline emerging in the post-genomic era that involves the comprehensive measurement of all metabolites in biological samples. Metabolites are substances, usually small molecules, produced by an organism when it breaks down food, drugs, chemicals, or its own tissues.

Currently, human non-targeted metabolomic studies have only identified about 10% of molecular features, which means that 90% of molecular features are still unknown.

In a new study published in Nature Biotechnology, a large international research team led by University of California, San Diego describes a new method of untargeted metabolomics that compares all metabolites in a sample with Matching of large sample databases involving lists of chemicals provides an unprecedented catalog of molecular signatures produced by food or food processed in the gut.

Untargeted mass spectrometry is a very sensitive technology that can detect hundreds or thousands of molecules in complex biological samples. In the new study, it could also be used to create individual dietary profiles. A deeper understanding of how the food we eat is converted into metabolites and by-products will have a direct impact on human health.

Study Corresponding author Dr. Pieter Dorrestein, director of the Center for Collaborative Mass Spectrometry Innovation at UC San Diego, said: "We can now link molecules in the diet to clinical health outcomes, not one at a time, but all at once. This was not possible before."

The study also used metagenomics to measure genetic material in biological samples and characterize the microorganisms present. The new method uses reference data-driven (RDD) analysis to match metabolomic data from tandem mass spectrometry (MS/MS) analyzes to metadata-annotated data.

Basically, each molecule has been stripped of electrons to make it electrically charged. The charged ions are then weighed using a very sensitive scale, which is then crushed into pieces before being weighed, creating a unique fingerprint for each molecule. Collections of these "fragment spectra" can be matched between the sample being analyzed and a reference database.

To exemplify RDD metabolomics, and because diet is critical to health, the team created a food metabolomics reference dataset. We then evaluated whether RDD metabolomics can accurately recall known food intake using dietary information from controlled participants in a study of sleep and circadian rhythms.

In the study, participants were put on a controlled diet twice over four days. So researchers can know how accurate the analysis results are. The results showed that 11 of the 15 foods directly matched the foods provided to the participants. Of the 11 foods, 3 matched fermented versions of non-fermented foods consumed, such as: fermented grapes (not grapes), cider (not apples), and yogurt (not milk). There were also 4 food categories that were not recorded during the study period. This demonstrates that RDD can successfully obtain correct historical dietary information from untargeted metabolomic data.

To evaluate whether RDDs can reveal dietary preferences, the team also analyzed a dataset of omnivores and vegetarians. Principal component analysis of relative proportions of reference foods revealed different patterns between dietary preferences: Omnivores had a higher match for dairy, meat, and seafood; vegetarians had a higher match for legumes, fruits, and vegetables . An RDD analysis of older adults found that individuals with lower dietary diversity were more likely to be matched to dairy, soda, and coffee, and that this diet type was more prevalent in the Alzheimer's group.This suggests that RDD analysis can be used to retrospectively stratify clinical studies based on empirical data on the dietary composition of each sample.

Dr. Rob Knight, co-corresponding author of the study and director of the Microbiome Innovation Center at the University of California, San Diego, said: "The new method will have a huge impact on Alzheimer's research because patients may not be able to remember or explain what they eat. stuff. In addition, this new method could also lead to new advances in wildlife conservation. After all, it is impossible for gorillas and cheetahs to fill out food questionnaires," the team said. , the molecular weights in blood or stool improved greatly when the food was matched to the group, so that, for example, a diet from Italy could be analyzed to match the centenarians from Cilento, Italy, who the UC San Diego team was studying. "This really shows how important it is to get food samples and clinical samples from around the world to allow us to understand how molecules and microbes interact with each other and improve or decrease our health outcomes based on specific diets," Knight said.

In summary, this New methods could provide a more complete understanding of the sources of many chemicals in human, animal and environmental samples. Potential applications of RDD metabolomics include understanding dietary and nutrient intake, exposure risk, medication use, environmental allergens, pollution studies, microbiome surveys, and personal care product tracking to inform potential exposure and health effects.