New AI Tool Improves Detection of Genetic Causes in Rare Disorders

Despite broad access to genomic sequencing, many patients remain undiagnosed, prolonging uncertainty and delaying targeted care. This prolonged “diagnostic odyssey” can stretch close to a decade and strain clinical resources. A new study shows how an artificial intelligence approach could shorten that path by prioritizing likely disease genes directly from patient findings.

Hebrew University of Jerusalem researchers developed EvORanker, an artificial intelligence (AI) algorithm that narrows thousands of candidate genes to the most probable disease-causing targets. The method compares genetic patterns across more than 1,000 species to detect evolutionary relationships between genes, including those not previously associated with human disease. By leveraging these cross-species signals, the tool surfaces hidden candidates that conventional knowledge-based pipelines may miss and ranks them for clinical review.

In clinical testing described by the team, EvORanker identified the correct causal gene as the top candidate in nearly 70% of cases and placed it within the top five in 95% of cases. The approach outperformed existing tools, particularly for poorly characterized genes. Case examples in the report include a child with a complex neurodevelopmental disorder in whom the algorithm highlighted a previously unrecognized gene, and another patient with a severe multisystem condition for whom the genetic basis was clarified, guiding next investigative steps.

The study is published in Genetics in Medicine on March 6, 2026. According to the institution, the work extends more than a decade of efforts that integrate evolutionary biology with computational analysis to accelerate gene discovery for undiagnosed patients. The researchers note that EvORanker is currently accessible to clinicians and scientists, with additional studies underway. They also report that rare diseases affect up to 5% of the global population and as much as 8% in Israel, underscoring the need for scalable, accurate gene prioritization. While rare disorders are the immediate focus, the team states the same strategy is being applied to oncology research to probe unexpected tumor regression and related mechanisms.

"There are thousands of cases like that around the world that fall through the cracks of current medicine," said Prof. Yuval Tabach, Faculty of Medicine, Hebrew University of Jerusalem. "Our goal was to give patients and clinicians a tool that can find fast and accurate answers where none existed before."

Related Links
Hebrew University of Jerusalem