We were particularly motivated to investigate the relationship of sentinel location to their target gene. This is because follow-up of GWAS hits typically proceeds by defining an associated region that spans the hit and is bounded by recombination hotspots on either side. All genes in this region are then explored for potential causality. We note that recombination hotspots vary in their strength, and this further adds to the arbitrary nature of this procedure. Furthermore, there is no biological reason for linkage disequilibrium patterns to have a direct bearing on the location of external regulators of transcription. To explore the validity of GWAS follow-up via associated intervals, we classified our cis-eQTL signals according to whether the sentinel marker resided inside (internal eQTL signal) or outside its target gene (external eQTL signal). External cis-eQTL signals were further classified according to whether the transcript being regulated was inside or outside the hotspot-bounded associated region of the marker. To account for uncertainty in the location of the true causal variant due to linkage disequilibrium, we redefined cis-eQTL signals according to the location of all
