Characterizing genetic interactions using a machine learning approach in Colombian patients with Alzheimer's disease

A main goal of human genetics is to understand the relationship between variations in DNA sequences and the susceptibility to certain illnesses. In this particular work, genetic information is analyzed in relation to the Alzheimer's disease (AD) in order to improve its diagnosis, prevention and treatment. In Colombia, this disease currently requires special attention because its incidence has increased significantly in recent years. Thus, this work analyzes a set of twelve genetic markers or single nucleotide polymorphisms (SNPs) in a set of Colombian patients through a constructive induction method based on a machine learning approach, namely, multifactor dimensionality reduction (MDR). Also, some statistical epistasis analysis is carried out. Particularly, epistasis is obtained based on information gain from AD related genes, providing a simple methodology to characterize interactions in genetic association studies and capturing important traits that describe the behavior of the disease.