It was long believed that the so-called FMR1 premutation — an excessive number of trinucleotide repeats (55-200 repeats; normal number of repeats is 5 to 40) in the FMR1 gene — had no direct effect on the people who carry it. Until recently, the only recognized effect on the carriers of the flawed gene was the risk of having offspring with fragile X syndrome (>200 of the trinucleotide repeats), a rare but serious form of developmental disability. In recent years, however, at least two clinical conditions have been well documented in the carriers themselves: an age-dependent neurodegenerative disorder and, in female carriers, early menopause. Now, a team of researchers from the University of Wisconsin (UW)–Madison and Wisconsin’s Marshfield Clinic has found that there may be a much broader health risk to carriers, with potentially dozens of clinical conditions that can be ascribed directly to carrying the premutation. The researchers employed machine learning, a form of artificial intelligence (AI), to mine decades of electronic health records (EHRs) of nearly 20,000 individuals in order to make this landmark discovery. In a study published online on August 21, 2019 in Science Advances, the team led by Marsha Mailick (photo), PhD, a researcher and Professor at UW–Madison’s Waisman Center, and Emeritus Vice Chancellor for Research and Graduate Education at UW-Madison, together with UW–Madison graduate student Arezoo Movaghar, provides a better understanding of the previously disputed relationship between this well-known genetic premutation and a wide range of clinical conditions. At the same time, the interdisciplinary study richly illustrates the power of data-driven discovery.
Login Or Register To Read Full Story