MicroRNA expression classification for pediatric multiple sclerosis identification

Abstract MicroRNAs (miRNAs) are a set of short non-coding RNAs that play significant regulatory roles in cells. The study miRNA data produced by Next-Generation Sequencing techniques can be valid help for the analysis multifactorial diseases, such as Multiple Sclerosis (MS). Although extensive studies have been conducted on young adults affected MS, very little work has done to investigate pathogenic mechanisms pediatric patients, and none from machine learning perspective. In this work, we report experimental results classification aimed at evaluating effectiveness methods automatically distinguishing MS healthy children, based their expression profiles. Additionally, since Attention Deficit Hyperactivity Disorder (ADHD) shares some cognitive impairments with also included patients ADHD our study. Encouraging were obtained an artificial neural network model features selected feature selection algorithms. show models developed overcome human experts. Developing automatic predictive support clinicians early diagnosis provide new insights find novel molecular pathways involved disease.