Gene expression analysis of pediatric Multiple Sclerosis using Machine Learning

Multiple Sclerosis (MS) is a chronic inflammatory demyelinating disease of the central nervous system that usually affects young adults. However, onset in childhood and adolescence is increasingly recognized by researchers, accounting for 3-5% of cases of MS. Improvements in diagnostic tools and/or greater sensitivity to early signs, in fact, have contributed to a better recognition of MS in the very early ages; while, in the past, it had mostly been monitored retrospectively. Genetic predisposition, environmental factors, and lifestyle appear to contribute significantly to the overall risk of developing MS; however, very few studies have investigated the "environmentally naïve" genetic load of pediatric MS (PedMS). Studying the transcriptomic involvement of PedMS can help elucidate the pathogenic mechanisms underlying MS in its early stages, which are not fully understood yet. The bioinformatic pipeline usually developed for differential gene expression analysis applies traditional statistical tests to search for genes that are differentially expressed between healthy controls and diseased patients. This analysis allows biologists to isolate evident changes in expression; however, it may fail to find more complex, nonlinear interactions among disease-related genes.