Machine Learning Predicts Age of Onset for Polyglutamine Diseases

Researchers at Niigata University in Japan have developed a model to predict the asymptomatic probability of polyglutamine diseases at each age from the current age and number of CAG repeats in carriers of spinocerebellar degeneration. Using machine learning, the team compared the predictive accuracy of two survival analyses with six parametric survival analyses. The two machine-learning methods showed a higher prediction accuracy than parametric survival analyses, with Random Survival Forests achieving the highest prediction accuracy. The researchers hope that the study will aid genetic counseling for career life planning and lead to more accurate predictions of the probability of disease onset. The results, Machine Learning Approach for the Prediction of Age-Specific Probability of SCA3 and DRPLA by Survival Curve Analysis, were published in Neurology Genetics.