Emulsifiers do the hard work of mixing oil and water so certain foods stay blended and smooth. Most of the emulsifiers we use today come from animals or synthetic chemicals.

Consumers want more natural, plant-based options, but finding the right plant proteins takes years of costly trial-and-error research.

A team from the University of Leeds and the University of Edinburgh built an AI tool to sort through tens of millions of plant proteins to find the ones that work.

“As we want to shift towards more sustainable, plant‑based ingredients, scientists face a major challenge: There are millions of potential plant proteins, but testing them all to identify the right emulsifier is expensive and involves a time‑consuming trial-and-error approach,” Dr. Simha Sridharan, who led the research, said. “Until now, there has been no reliable way to predict which plant proteins are likely to behave as emulsifiers like animal proteins.”

How AI Sorts Plant Proteins

Advertisement

Photo: earthphotostock/Shutterstock

The team used a simulation model to see exactly how proteins attach themselves at the interface between oil and water. They combined statistical physics and machine learning to “fingerprint” specific parts of the proteins to see how they behave.

“Emulsifiers often have a characteristic chemical structure called di-blocks,” Dr. Rik Sarkar, from the University of Edinburgh, explained. “We were able to model this structure mathematically for plant proteins. Using machine learning based on features obtained from statistical physics simulations, we can predict which plant proteins are most likely to work best as natural emulsifiers.”

Real Testing

Out of millions of options, the AI found nearly 800 promising plant proteins. These impressive numbers cut years of expensive lab testing in a fraction of the time.

“The model identified nearly 800 plant proteins that could potentially act as emulsifiers, many of which had never previously been considered for this purpose,” Professor Anwesha Sarkar shared. “We then tested several commercially available proteins and found the results matched the model’s predictions, with proteins from peas and potatoes proving effective. This shows how AI could help researchers find promising new ingredients much faster than before.”