Human QTL studies aim to understand the molecular mechanisms underlying individual variation in biological quantitative traits, many of which are established risk factors for complex common diseases (de Koning and Haley, 2005). Traditional approaches to QTL mapping in humans have largely focused on the identification of loci affecting just one or a few biological quantitative traits (Kendziorski et al., 2006). However, current microarray technologies allow measurement of many thousand gene expression levels each of which are also quantitative traits that have been shown to be highly variable and heritable (Cheung et al., 2005). A focus of interest in QTL mapping is now shifting to a more systematic and comprehensive characterization of quantitative biological variation from the perspective of the genome itself. The exponential model of QTL effect sizes (based on Fisher's infinitesimal model of adaptive evolution) predicts that it should be possible to explain a substantial amount of quantitative genetic variation by a limited number of genes with large effect sizes. The variants that affect gene expression (expression QTL, eQTL) are thereby expected to have an important role in generating quantitative biological variation (Farrall, 2004).
The design of studies to identify these eQTL is similar to traditional QTL studies in that they aim to identify the locations in the genome to which the expression traits are linked (Kendziorski et al., 2006). Early experiments that combined expression profiling with classical genetic mapping approaches, such as the studies of Cheung et al. (2005), have revealed a wealth of ''expression phenotypes'' in the human genome. These eQTL studies which investigate thousands of expression transcripts face significant challenges in statistical inference (Kendziorski etal., 2006) and the need for new computational resources to visualize and explore data from combined genome-wide expression and linkage studies (Carlborg et al., 2005; Mueller et al., 2006).
Was this article helpful?