Identification of Rare Genetic Disorder from Single Nucleotide Variants Using Supervised Learning Technique

Abstract

Muscular dystrophy is a rare genetic disorder that affects the muscular system which deteriorates the skeletal muscles and hinders locomotion. In the finding of genetic disorders such as Muscular dystrophy, the disease is identified based on mutations in the gene sequence. A new model is proposed for classifying the disease accurately using gene sequences, mutated by adopting positional cloning on the reference cDNA sequence. The features of mutated gene sequences for missense, nonsense and silent mutations aims in distinguishing the type of disease and the classifiers are trained with commonly used supervised pattern learning techniques.10-fold cross validation results show that the decision tree algorithm was found to attain the best accuracy of 100%. In summary, this study provides an automatic model to classify the muscular dystrophy disease and shed a new light on predicting the genetic disorder from gene based features through pattern recognition model.