Using AI to Diagnose Rare Diseases through Real-World Symptoms

Rare Daily Staff

Researchers at King Abdullah University of Science and Technology have developed an artificial intelligence tool to diagnose patients with rare diseases by leveraging a diverse range of data sources to identify genetic variants associated with diseases.

Dubbed STARVar–Symptom-based Tool for Automatic Ranking of Variants, it is an AI-powered approach that uses background information from scientific literature, genomic information from DNA sequence reads, and clinical symptoms from individual patient records to make a diagnosis.

The researchers say the new AI tool is distinguished from other gene prioritization approaches because of its focus on real-world patient symptoms, regardless of how these clinical descriptions are documented.

Traditional methods often demand that clinical presentations adhere to standardized vocabularies, impeding a more nuanced and accurate understanding of patient symptoms. The researchers said the reality, though, is that doctors and researchers frequently convey patient data using terminology that extends beyond predefined terms.

“STARVar stands a unique and efficient tool that has the advantage of prioritizing genomic variants by using flexibly expressed patient symptoms in free-form text,” said Șenay Kafkas, a bioinformatics researcher at King Abdullah University of Science and Technology (KAUST) and the first author of a report that details the tool.

Designed by KAUST computer scientist Robert Hoehndorf and members of his team, the method can interpret symptom data recorded in either standardized or natural language formats.

When evaluated on different genomic datasets—generated using clinical variants collected from patients, both in Saudi Arabia and from other countries around the world—STARVar outperformed several other variant prioritization tools that can operate with only rigidly represented symptoms. In particular, the algorithm consistently ranked the correct disease-associated variant at or near the top of the list of potential candidate variants in these validation tests.

In one example, out of nearly 800 suspect gene variants uncovered by genomic sequencing, STARVar narrowed down the possibilities to a single mutation. This mutation in the MMP2 gene was already known to be pathogenic and was implicated as the likely driver of the girl’s condition.

STARVar is now available online. Kafkas hopes to see the clinical genetics community embracing it and integrating the analytic method into their genomic workflows. “STARVar stands as a unique and efficient tool,” she said, “one that will shed light on rare diseases and provide vital diagnostic support to clinicians and affected families.”

Photo: Șenay Kafkas and Marwa Abdelhakim who developed STARVar