Nonparamatric Tests Of Association With Rare Variant Based On Haplotype

Since human complex diseases or common diseases are usually caused by both common and rare variants.In recent years,Genome-wide association studies(GWAS)have been successful in detecting common variants,which are associated with complex and common diseases.However,recent studies have demonstrated that common variants can explain only a small proportion of heritability,more complex diseases have a strong correlation with rare variants.Recent genome-wide association studies(GWAS)faces great challenges in understanding the impact of rare variants,and most existing methods are detecting the association between common variants and complex diseases.Therefore,how to propose an efficient method to detecting rare variants is a hotspot issue in current society.Association analysis is an effective way in investigating genetic variations associations with human diseases,and haplotype can provide insights into underlying linkage disequilibrium(LD)structure.In the existing methods of detecting rare variants,association analysis based on haplotype is considered to have outstanding advantages over other methods.In this article,we first introduce several association analysis methods based on haplotype,then,we also introduce a nonparametric test association with rare variants based on haplotype,and proposed a new statistic.The asymptotic distribution of this statistic is not readily available because of the rare variants.So we use permutation test.At the same time,our method overcomes some common problems,such as the large degrees of freedom,too many parameter and so on.From the results of simulation,at the given level of significance,we found that the type one error of the proposed method is reasonable.We also compare our methods with HKAT,Wei-SIMc-matching,SKAT and SKAT-O.In addition,we aslo apply the method proposed in this paper to the hypertension data provided by WTCCC,analysis the correlation between gene ZFAT1 and hypertension.The results show that our method is powerful than other methods.So,considering the power and robustness of the method,and the feasibility of calculation,we recommend using the method proposed in this article to detect rare variants.