Skip to main content

A group of researchers from different countries has reported a new technique that can identify many molecules from food that could not previously be detected in stool or blood samples.

The study team, led by researchers at the University of California, San Diego, successfully matched each metabolic product in a specimen to existing reference databases using the method known as metabolomics. not targeted. He said the approach provides an “unprecedented catalog of molecule signatures” that are formed from eating or processing food in the gut.

Junk food

“Untargeted mass spectrometry is a very sensitive technique that allows the detection of hundreds to thousands of molecules that can now be used to create a diet profile of individuals,” said Pieter Dorrestein, Ph.D. UC San Diego, corresponding author of the research.

This new method could significantly expand knowledge about the origins of chemicals seen in many human, animal and environmental samples, the researchers said. It makes it possible to link, for the first time, molecules in food to health or clinical results.

The method was reported in the July 2022 issue of Nature Biotechnology.

Search for metabolites

Metabolomics is the study of metabolites, which are chemical substances resulting from metabolism. In other words, metabolites are substances that emerge from the breakdown of foods, drugs, chemicals or tissues.

An emerging field of study, metabolomics involves the identification and measurement of metabolites in a biological system. Existing research has succeeded in identifying only a small fraction (about 10%) of the molecular characteristics of sampled specimens.

Using reference data-based (RDD) analysis, this new method compares metabolomic data from tandem mass spectrometry (also known as MS/MS) to data from a pseudo-MS/MS reference library.

MS/MS is an instrumental analysis method that uses two analyzers, rather than one, to measure molecular weight.

Scientists have stripped molecules of electrons to make them charged. They weighed the charged ions with a very sensitive balance, then broke them into pieces, which were also weighed. This process creates a unique fingerprint for each of these molecules.

Researchers match the resulting coin pools between analyzed samples and reference databases. So far this has been quite difficult to do.

Another related method known as metagenomics was also used in the study. This helps measure genetic material derived directly from biological or environmental samples and describe the microbes they contain.

Find out what you eat

Scientists in this new research examined thousands of foods to identify the molecules present. These foods were submitted by people around the world as part of the Global FoodOmics project, launched seven years ago at UC San Diego.

The data output from using the new approach was more than five times that of traditional methods. The most interesting part is that the new technique allows the use of off-target metabolomics to determine diet using blood or stool samples.

RDD analysis allowed researchers to explore dietary patterns — for example, vegan versus omnivorous — and consumption of certain foods. This allowed the team to match metabolomics data with existing reference databases.

“The ability to directly read diet from a sample has huge implications for research on populations like people with Alzheimer’s disease, who may not be able to remember or explain what they ate,” said co-corresponding author Rob Knight, Ph.D. “And in wildlife conservation applications. Good luck finding a cheetah or a gorilla, to name just two of the hundreds of species we’re studying, to fill out a food diary.

The researchers noted that this work highlights the importance of obtaining food specimens and clinical samples from around the world. This would help to better understand how molecules and microbes combine to improve or worsen health based on diet.


Improving off-target metabolomics using metadata-based source annotations