One salient advantage of iPSC-based regulatory variation studies such as eQTL/ASE studies is that candidate functional genes and variants can be rigorously tested and validated within the same system (i.e., pluripotent stem cells and differentiated cells), one that is readily amenable to CRISPR-Cas9 genome editing for the purpose of studying gene expression (Soldner et al., 2016). This cannot easily be done with findings from human tissues such as whole liver, except in unusual cases where there is a high degree of orthology between the corresponding loci in human and an appropriate animal model (such as mouse) that allows for genome editing of a conserved candidate noncoding variant in the animal model in vivo (as was the case with rs2277862 and rs27324996 in the CPNE1/Cpne1 loci). We were able to rigorously assess for functionality of the MPRA-nominated rs2277862, rs10889356, and rs10872142 through the generation of isogenic wild-type and variant hPSCs via genome editing. Consistent with prior studies, we found the generation of knock-in clones via HDR to be very inefficient compared to the generation of clones with defined deletions via multiplexed
