eQTL-based analyses represent one strategy to detect how risk alleles exert their effects on gene expression. False negative results can be attributed to biologic and technical reasons. For example a variant may influence expression only at a certain developmental timepoint or in a non-cell autonomous fashion. In addition, most of the eQTL analyses are based on measures of steady-state expression levels, how ever a variant could be influencing the non-steady-state aspect of expression, such as rate of transcription. A negative eQTL result may also arise from technical limitations, which can occur due to a lack of power to detect a subtle, but biologically important change in transcript level. Testing associations with other transcribed elements, such as transcript isoforms as well as small and long non-coding RNAs will also be informative. Moreover, it will be interesting to compare eQTL-target gene results in tumors with results derived from normal tissues. The ability to systematically test many of these hypotheses in the near future will be possible given the proliferation of deeply characterized datasets (e.g., GTEx –http://commonfund.nih.gov/GTEx/;MuTHER -http://www.muther.ac.uk/).
