Analysis of SNP-Complex Disease Association by a Novel Feature Selection Method

Selecting a subset of SNPs (Single Nucleotide Polymorphism pronounced snip) that is informative and small enough to conduct association studies and reduce the experimental and analysis overhead has become an important step toward effective disease-gene association studies. In this study, we developed a novel methods for selecting Informative SNP subsets for greater association with complex disease by making use of methods of machine learning. We constructed an integrated system that makes use of major public databases to prioritize SNPs according to their functional effects and finds SNPs that are closely associated with genes which are proven to be associated with a particular complex disease. This helped us gain insights for understanding the complex web of SNPs and gene interactions and integrate as much as possible of the molecular level data relevant to the mechanisms that link genetic variation and disease. We tested the validity and accuracy of developed model by applying it to real life case control data set and got promising results. We hope that results of this study will support timely diagnosis, personalized treatments, and targeted drug design, through facilitating reliable identification of SNPs that are involved in the etiology of complex diseases.