Scientists at the University of Leeds have created an artificial intelligence (AI) method that can quickly identify plant proteins that act as emulsifiers—substances that help mix oil and water into stable, uniform solutions. This discovery, published in Communications Chemistry, could significantly cut the time and cost of developing plant-based foods and cosmetics, which are becoming increasingly popular as alternatives to traditional animal-based or synthetic ingredients. Emulsifiers are found in a wide range of products, including lotions, ice cream, mayonnaise, and paints, where they help maintain a smooth and consistent texture. The research, led by Dr. Simha Sridharan and supervised by Professor Anwesha Sarkar from the University of Leeds, involved collaboration with AI experts at the university and Dr. Rik Sarkar, a machine learning specialist at the University of Edinburgh. Scientists previously faced a major challenge in identifying which plant proteins could function as emulsifiers, as testing each one individually was both costly and time-consuming. There was no reliable method to predict which proteins would behave like the commonly used animal-derived emulsifiers, such as casein or whey proteins found in milk. To overcome this, the team used a simulation model to understand how proteins interact at the interface between oil and water. They then applied machine learning to identify specific segments of the proteins that influence this interaction. By combining machine learning with statistical physics, the researchers were able to screen a vast number of plant proteins and predict which ones would perform similarly to animal-based emulsifiers. This process significantly reduced the time required for traditional testing methods. The model identified nearly 800 plant proteins that could potentially act as emulsifiers, many of which had not previously been considered for this role. Early tests confirmed the accuracy of the predictions, with proteins from peas and potatoes showing effective emulsifying properties. This breakthrough could help food and cosmetics companies develop sustainable, plant-based products more efficiently. Dr. Rik Sarkar noted that emulsifiers often have a specific chemical structure called "diblocks," and the team successfully modeled this structure for plant proteins using machine learning and physics-based simulations. This research highlights the growing role of AI in accelerating scientific discoveries and advancing the development of eco-friendly alternatives.