For patients with rare diseases, getting a diagnosis can take years. Even with modern genetic testing, many cases remain unresolved, leaving families without answers and doctors without a clear path forward. A newly developed AI tool called EvORanker, created by researchers at the Hebrew University of Jerusalem, aims to change that.
What makes EvORanker different is its data source. Rather than drawing solely on existing medical knowledge, it analyses how genes have evolved across more than 1,000 species. By identifying patterns of co-evolution, the system can detect functional links between genes, including genes not previously connected to any known disease.
In clinical testing, published in Genetics in Medicine, the algorithm correctly identified the disease-causing gene as its top candidate in nearly 70% of cases, and placed it within the top five candidates in 95% of cases. In one example, it pinpointed the genetic cause of a child's neurodevelopmental disorder after conventional testing had drawn a blank. In another, it uncovered the genetic basis of a severe multisystem disease, opening potential treatment options.
Beyond diagnosis, the tool may also help match patients to existing drugs that target newly identified disease genes, a process known as drug repurposing. Researchers are already exploring whether EvORanker can be extended to cancer, where understanding genetic behaviour could improve treatment decisions.
The tool is now available to clinicians and researchers, with further validation studies underway. Rare diseases affect an estimated 5% of the global population, and for many of those patients, a faster diagnosis could be life changing.
References:
ICT&health: AI tool may shorten the diagnostic journey for rare diseases
