Sequencing or genotyping the genome of individuals allows the characterization of what distinguishes the hereditable material of each individual from that of others. By matching differences in phenotype (e.g., blood pressure, adult height) between individuals and this genomic characterization (i.e., in association studies) genetic epidemiologists are able to impute these phenotypic differences to a small span of the genome. The smallness of the span is a function of the spatial resolution with which the genomic characterization occurs. Although there is controversy around what constitutes sufficient resolution, there is some consensus that genomic markers such as SNPs spaced every thousand bases will be sufficient to unambiguously resolve the span of the genome associated with a phenotypic difference to a single gene . Currently, the cost of a single genotype is around $0.50 and so the cost of a high-resolution genome scan of an individual is on the order of magnitude of $1 million. At this cost, only a very few institutions can afford a comprehensive study of a population. If the recent past is to be a guide, the cost of genotyping is likely to drop by several orders of magnitude well within a decade, at which point genome-wide scans of populations will become economically feasible.
The kind of information that these studies will provide includes the contribution of particular polymorphisms to changes in phenotype, presumably via changes in gene function, pattern of expression, or both. Currently, expression microarrays capture information about gene polymorphism poorly, if at all. Presently shipping microarrays typically code for a canonical "normal" gene sequence. Departures from this sequence will result in changes in the intensity reading reported by any of the currently employed microarray platforms, as will become obvious after reading the chapter on measurement techniques (chapter 3). Undoubtedly, in the next 5 years, the continued geometric increase in the complexity and density of expression microarrays will allow specific assaying for all common and all clinically significant polymorphisms. Until then, however, polymorphisms will be a source of "noise" or unexplained variation. Functional genomics investigations designed to elucidate the role of particular polymorphisms in phenotypic mechanisms or phenotypic variation are best conducted using methods other than RNA expression detection microarrays, such as high-throughput sequencing and genotyping. These latter techniques are beyond the scope of this text (see [43, 112] for an excellent survey of sequencing and genotyping) and therefore this particular avenue of functional genomics will not be addressed below except tangentially.
Was this article helpful?