Finally, methods for integrating different types of association signals are developing. A nascent view proposes that combining genome-wide expression and genotyping data into a joint quantitative signal can increase power for discovery [6, 37, 77, 78]. One particularly attractive feature of this view is that it augments structure (genotype) with function (expression). Indeed, one study demonstrated that SNPs correlated with gene expression changes (expression quantitative trait loci = eQTLs) were more likely to show disease association than other SNPs from a GWAS array [79]. Relatedly, visualization tools can graphically overlay association metrics onto other data in order to prioritize markers. Visualization has been used to integrate SNP association with quantitative imaging phenotypes [80], among other examples.
