Preventing food spoilage to feed more people

Benedetto Marelli (MIT), Markus Buehler (MIT), Pin-Yu Chen (IBM), and Lingfei Wu (IBM)


Food spoilage is a global threat to food safety and security; roughly one-third of the food produced worldwide for human consumption is lost or wasted, and almost half of all the fruit and vegetables produced are wasted. MIT-IBM researchers are searching for new materials to preserve food and prevent waste due to spoilage. Saving even one-fourth of the food currently lost or wasted globally could help feed 870 million hungry people. The scientists will develop a system to design, build, and test new structural proteins that could be used as edible coatings to extend the shelf life of perishable foods. First, they will use machine learning techniques to analyze the amino acid sequences that make a protein edible, nontoxic, stable, and preservative. Then, they’ll apply AI algorithms to model the three-dimensional structures of the novel predicted proteins and predict their function. Structural modeling is a time- and resource-intensive process that can be accelerated by AI. Finally, they plan to synthesize the best candidates in a lab to validate the algorithms’ predictions. Their system could also be used to identify and develop new materials for other purposes, such as novel drugs.