Random Forest and Gene Networks for Association of SNPs to Alzheimer's Disease

Machine learning methods, such as Random Forest (RF), have been used to predict disease risk and select a set of single nucleotide polymorphisms (SNPs) associated to the disease on Genome-Wide Association Studies (GWAS). In this study, we extracted information from biological networks for selecting candidate SNPs to be used by RF, for predicting and ranking SNPs by importance measures. From an initial set of genes already related to a disease, we used the tool GeneMANIA for constructing gene interaction networks to find novel genes that might be associated with Alzheimer's Disease (AD). Therefore, it is possible to extract a small number of SNPs making the application of RF feasible. The experiments conducted in this study focus on investigating which SNPs may influence the susceptibility to AD.