Researchers at the Department of Neurology, Niigata University used machine learning to develop a model that predicts the probability of being asymptomatic at each age from the current age and number of CAG repeats in carriers of spinocerebellar degeneration.
Polyglutamine diseases such as DRPLA and SCA3 are caused by expansion of CAG repeats in causative genes. It is known that the number of CAG repeats is inversely proportional to the age of onset in polyglutamine diseases. Parametric survival analysis has traditionally been used to predict age of onset, but more accurate prediction methods have been desired.
Researchers used two machine learning survival analyzes to predict age of onset and compared their accuracy to six parametric survival analyzes. Two machine learning methods (Random Survival Forest and DeepSurv) showed higher prediction accuracy than parametric survival analysis. In particular, Random Survival Forests had the highest prediction accuracy and was used for the final prediction.
“This study is important for genetic counseling for career and life planning. Going forward, we will continue to analyze more cases at several institutions with the aim of more accurately predicting the probability of developing the disease.” said Hatano and Dr. I’m Ishihara.
Research results “A machine learning approach for predicting age-specific probabilities of SCA3 and DRPLA by survival curve analysis” were published in the online edition of the journal. neurology genetics.
For more information:
Yuya Hatano et al, A machine learning approach for prediction of age-specific probabilities of SCA3 and DRPLA by survival curve analysis, neurology genetics (2023). DOI: 10.1212/NXG.0000000000200075
Provided by Niigata University
