COMPUTATIONAL TOOLS FOR COMPLEX TRAIT GENE MAPPING

摘要

The mapping of the genes underlying complex traits poses special challenges. The results of several years of effort by many groups in the extension of the linkage mapping methods, used with great effect for localizing major genes, has been disappointing on the whole for complex traits. Now that we have an effectively complete genome sequence and exciting new technologies for genotyping vast numbers of single nucleotide polymorphisms (SNPs) the way is open for the advance of a new strategy. There have already been several successful outcomes for complex trait mapping through the analysis of linkage disequilibrium (LD) and haplotypes. However, these are early days and some of the difficulties are only slowly becoming apparent. Recent evidence suggests that the human genome may contain up to 15 million SNPs. For this reason the probability of actually including a disease causal SNP in a sample of SNPs typed at a spacing of several kilobases is low. Furthermore, this implies that up to 100 other SNPs may be in linkage disequilibrium with a causal SNP. This poses major difficulties for identifying a causal site but the initial target is simply to determine candidate regions with confidence. The International HapMap project has the aim of delimiting haplotype blocks in a number of populations to generate a genome-wide SNP map for association studies. One outcome of this project will be a large body of empirical data on patterns of linkage disequilibrium across the human genome. Other groups and organizations are involved in their own data collection and evaluation studies. Aspects of the effective collection, repiesentation and use of these vast and developing data resources are the topics of the six papers included in this volume.