Scientists from Harvard Medical School in the United States and the Centre for Genomic Regulation in Barcelona, Spain, have developed an artificial intelligence (AI) model capable of pinpointing the mutations in human proteins most likely to cause disease, according to a study published on November 24 in Nature Genetics.

The model, named popEVE, represents a potential breakthrough in the diagnosis of genetic disorders. Researchers said it leverages evolutionary data from hundreds of thousands of species along with genetic variation information from human populations worldwide.

By analyzing this extensive evolutionary record, popEVE can map critical and variable regions across more than 20,000 human proteins. This allows the tool not only to identify disease-causing mutations but also to rank them according to the level of harm they may pose to the human body, offering a more precise approach to understanding genetic risk.

The developers said popEVE could transform how genetic diseases are diagnosed and studied, potentially improving early detection and treatment strategies.