A new multipoint method for genome-wide association studies by imputation of genotypes英文版本.pdf

T E C H N I C A L R E P O R T S s c i A new multipoint method for genome-wide association t e n e g studies by imputation of genotypes e r u t a n 1,2 1,2 1 1 1 / Jonathan Marchini , Bryan Howie , Simon Myers , Gil McVean Peter Donnelly m o c . e r u t Genome-wide association studies are set to become the first of these concerns, but there is currently no consensus on the a n w. method of choice for uncovering the genetic basis of human best approach. w diseases. A central challenge in this area is the development of In this paper, we suggest a coherent framework for thinking about w / / powerful multipoint methods that can detect causal variants this problem and then illustrate this with a number of different : p t that have not been directly genotyped. We propose a coherent applications. The main idea behind our approach is to think of the t h analysis framework that treats the problem as one involving problem as one involving a combination of observed data and missing p u missing or uncertain genotypes. Central to our approach is a data, where the core aim is to predict (or ‘impute’) the missing o r model-based imputation method for inferring genotypes at data based upon the observed data. All multipoint methods can be G g observed or unobserved SNPs, leading to improved power over thought of in terms of this prediction aspect, but many are not n i existing methods for multipoint association mapping. Using