A newly developed artificial intelligence (AI) tool could significantly reduce the time needed to identify the genetic causes of rare diseases, offering new hope to patients and families facing years of uncertainty. The research was led by scientists at the Hebrew University of Jerusalem.
For many patients with rare diseases, obtaining a diagnosis can take years, often referred to as a “diagnostic odyssey.” Despite advances in genetic sequencing, a large proportion of patients remain undiagnosed, leaving clinicians without clear guidance for treatment. The newly developed AI algorithm, called EvORanker, aims to address this challenge by identifying which gene, among thousands of candidates, is most likely responsible for a patient’s condition.
Evolutionary data as a new diagnostic source
Unlike conventional tools that rely primarily on existing medical knowledge, EvORanker analyses how genes have evolved across more than 1,000 species. By detecting patterns of co-evolution, the system can uncover functional relationships between genes, including those not previously associated with disease.
In clinical testing, the algorithm identified the correct disease-causing gene as the top candidate in nearly 70 percent of cases and ranked it within the top five in 95 percent of cases. According to the researchers, this represents a significant improvement over existing diagnostic tools, particularly in complex cases involving poorly understood genes. The research was recently published in Genetics in Medicine.
Patient impact
The study highlights several clinical examples. In one case, the algorithm identified a previously unknown gene linked to a child’s neurodevelopmental disorder after conventional diagnostics had failed. In another, it revealed the genetic basis of a severe multisystem disease, providing both diagnostic clarity and potential therapeutic direction.
“There are thousands of cases like that around the world that fall through the cracks of current medicine,” said Prof. Yuval Tabach. “Our goal was to give patients and clinicians a tool that can find fast and accurate answers where none existed before.”
Broader implications
Beyond diagnosis, the tool may also support drug repurposing by linking newly identified disease genes to existing therapies. This could accelerate treatment development and reduce time to clinical application. While the current focus is on rare diseases, which affect up to 5 percent of the global population, the researchers are already exploring applications in oncology, including understanding tumor behavior and treatment response.
EvORanker is now available to researchers and clinicians, with further studies underway to validate its clinical impact. If successful, the technology could help transform rare disease diagnostics by turning years of uncertainty into faster, data-driven answers.
