TY - JOUR
T1 - Improved power by use of a weighted score test for linkage disequilibrium mapping
AU - Wang, Tao
AU - Elston, Robert C.
N1 - Funding Information:
We thank the reviewers for their helpful comments, which improved the manuscript. This work was supported in part by U.S. Public Health Service resource grant RR03655 from the National Center for Research Resources, research grant GM28356 from the National Institute of General Medical Sciences, and cancer center support grant P30CAD43703 from the National Cancer Institute.
PY - 2007/2
Y1 - 2007/2
N2 - Association studies offer an exciting approach to finding underlying genetic variants of complex human diseases. However, identification of genetic variants still includes difficult challenges, and it is important to develop powerful new statistical methods. Currently, association methods may depend on single-locus analysis-that is, analysis of the association of one locus, which is typically a single-nucleotide polymorphism (SNP), at a time-or on multilocus analysis, in which multiple SNPs are used to allow extraction of maximum information about linkage disequilibrium (LD). It has been shown that single-locus analysis may have low power because a single SNP often has limited LD information. Multilocus analysis, which is more informative, can be performed on the basis of either haplotypes or genotypes. It may lose power because of the often large number of degrees of freedom involved. The ideal method must make full use of important information from multiple loci but avoid increasing the degrees of freedom. Therefore, we propose a method to capture information from multiple SNPs but with the use of fewer degrees of freedom. When a set of SNPs in a block are correlated because of LD, we might expect that the genotype variation among the different phenotypic groups would extend across all the SNPs, and this information could be compressed into the low-frequency components of a Fourier transform. Therefore, we develop a test based on weighted Fourier transformation coefficients, with more weight given to the low-frequency components. Our simulation results demonstrate the validity and substantially higher power of the proposed method compared with other common methods. This method provides an additional tool to existing methods for identification of causative genetic variants underlying complex diseases.
AB - Association studies offer an exciting approach to finding underlying genetic variants of complex human diseases. However, identification of genetic variants still includes difficult challenges, and it is important to develop powerful new statistical methods. Currently, association methods may depend on single-locus analysis-that is, analysis of the association of one locus, which is typically a single-nucleotide polymorphism (SNP), at a time-or on multilocus analysis, in which multiple SNPs are used to allow extraction of maximum information about linkage disequilibrium (LD). It has been shown that single-locus analysis may have low power because a single SNP often has limited LD information. Multilocus analysis, which is more informative, can be performed on the basis of either haplotypes or genotypes. It may lose power because of the often large number of degrees of freedom involved. The ideal method must make full use of important information from multiple loci but avoid increasing the degrees of freedom. Therefore, we propose a method to capture information from multiple SNPs but with the use of fewer degrees of freedom. When a set of SNPs in a block are correlated because of LD, we might expect that the genotype variation among the different phenotypic groups would extend across all the SNPs, and this information could be compressed into the low-frequency components of a Fourier transform. Therefore, we develop a test based on weighted Fourier transformation coefficients, with more weight given to the low-frequency components. Our simulation results demonstrate the validity and substantially higher power of the proposed method compared with other common methods. This method provides an additional tool to existing methods for identification of causative genetic variants underlying complex diseases.
UR - http://www.scopus.com/inward/record.url?scp=33846565755&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33846565755&partnerID=8YFLogxK
U2 - 10.1086/511312
DO - 10.1086/511312
M3 - Article
C2 - 17236140
AN - SCOPUS:33846565755
SN - 0002-9297
VL - 80
SP - 353
EP - 360
JO - American Journal of Human Genetics
JF - American Journal of Human Genetics
IS - 2
ER -