Predicting muscular dystrophy with sequence based features for point mutations

Hefty amounts of biological data are accumulated for research with the advancement of sequencing technologies. Genetic diseases are caused by the deformity in the inherited genes. Identifying trait diseases through DNA analysis is a prime task in diagnosing an ailment. Identification of disease based on mutations in the gene sequences is an essential and challenging task in the medical diagnosis of genetic disorders such as Muscular dystrophy. Muscular dystrophy is a rare disease that alters the structure and nature of the muscles that deteriorate the musculoskeletal system and hinder locomotion. There are nine major kinds of muscular dystrophy and it is vital to identify the type of muscular dystrophy for proper diagnosis and treatment. Hence a new model is proposed for predicting the disease accurately with the gene sequences, which are mutated by adopting an approach like positional cloning on the reference cDNA sequence. This paper addresses the problem by considering mutated gene sequences of fifty five genes that causes five types of muscular dystrophy and developing an efficient pattern recognition model using supervised pattern classification technique. The resultant the trained model shows the prediction accuracy of 100% by estimating using 10-fold cross validation.