An Application Of Tabu Table Based Negative Feedback Ant Colony Optimization Algorithm In Genome-wide Association Analysis

In recent years,with the completion of the Human Genome Project(HGP),more and more researchers have used genomic wide association study(GWAS)to study the genetic laws of many simple diseases.It provided help on treatment and prevention of diseases.GWAS finds single nucleotide polymorphisms(SNPs)related to diseases by detecting the association between SNPs and diseases.However,as people's living standards has been gradually improved,many complex diseases become common diseases,not caused by a single SNP mutation,but caused by multiple SNPs.Researchers used various methods to find combinations of pathogenic SNPs.In this thesis,we present a new algorithm,a negative feedback ant colony optimization algorithm based on a tabu list for GWAS.The advantage of this algorithm is that it uses mutual information and Gini score,which are relatively small computational objective functions to evaluate SNP.The degree of association between the combination and the disease greatly speeds up the calculation.And it search for the mutual information and Gini score SNP combinations with strong disease correlation,and then merges them.Using a dynamic threshold,most of the artificial ants in the initial stage of the algorithm enter negative feedback of random search,and a tabu list is added in the negative feedback stage to force the artificial ants to perform a global search and increase the spatial search capability of this algorithm.The algorithm in this paper is tested on the simulation data set,which not only ensures the accuracy,but also improves the calculation speed and saves time.In this paper,the algorithm is applied to the elderly macular degeneration data set,and a validated SNP is found.