![](https://medcitynews.com/uploads/2021/07/GettyImages-1280895783-150x150.jpg)
A new study out of Vanderbilt University Medical Center found genetic data in electronic health records can be used to spot undiagnosed diseases.
The research, which was recently published in Science, was authored by 27 individuals.
They believed people diagnosed with conditions like infertility, kidney failure, stroke and heart failure could actually have a rare genetic disease.
Thus, the researchers assigned scores to 21,701 people depending on how well their symptoms fit with a description of 1,204 genetic diseases. Patients included were taken from BioVU, a collection that links DNA samples to de-identified EHRs. The researchers replicated the results at a Marshfield Clinic biobank.
Overall, the study found 18 associations between genetic variants and high phenotype risk scores.
The researchers also learned that 14 percent of individuals with genetic variants impacting the kidney had kidney transplants. Ten percent with another variant had liver transplants. These transplants could have been avoided had the underlying genetic cause been disclosed. Instead, patients could have undergone another treatment to prevent the symptoms from worsening.
“We started with a simple idea: look for a cluster of symptoms and diseases to find an undiagnosed underlying disease,” Josh Denny, a professor of biomedical informatics and medicine, director of VUMC’s Center for Precision Medicine and a study author, said in a statement. “Then we got really excited when we saw how we could systemize it across thousands of genetic diseases to figure out the impact of millions of genetic variants.”
The method was developed by Denny; Lisa Bastarache, lead data scientist at the Center for Precision Medicine; and other team members.
Bastarache noted that this technique presents numerous chances for more research and findings.
“Phenotype risk scoring can easily be applied in any electronic medical record system that is linked to DNA,” she said in a statement. “Our work looked only at a small sample of the human genome, about 6,000 variants. The opportunity for additional discoveries using this method is huge.”
Photo: BlackJack3D, Getty Images