New Artificial Intelligence Model Could Speed Rare Disease Diagnosis

Rare Daily Staff

A new artificial intelligence model called popEVE can help doctors spot harmful genetic changes hidden among the tens of thousands of mutations in a person’s DNA.​

Developed by researchers at Harvard Medical School and other institutions, the model predicts whether genetic variants are harmless or disease-causing, and whether they are more likely to lead to death in childhood or adulthood, according to a paper published in Nature Genetics.

PopEVE flagged variants in 123 genes linked to developmental disorders that had not been previously reported, essentially uncovering likely genetic causes of those conditions. In 25 of those genes, later studies from other laboratories have independently confirmed that the variants can cause disease.​

The team hopes popEVE will help clinicians diagnose single-variant genetic diseases — especially rare diseases — more quickly and accurately, and that it could also guide the search for new drug targets.

Scientists can access popEVE through an online portal.​

“Our goal was to develop a model that ranks variants by disease severity, providing a prioritized, clinically meaningful view of a person’s genome,” said co-senior author Debora Marks, a professor of systems biology in the Blavatnik Institute at Harvard Medical School.