In addition, we employed the alternative approach of CRISPR interference, a term that is typically used to describe the repression of transcription via the positioning of dCas9 to a gene promoter or coding sequence to block transcriptional initiation or elongation via steric interference (Qi et al., 2013); alternatively, dCas9 can be attached to a repressor domain such as Krüppel associated box (KRAB) to effect transcriptional silencing via epigenetic chromatin modification (Gilbert et al., 2013). We opted to use the unadorned dCas9 protein rather than attaching an extra domain, since the regulatory elements in the vicinity of the three SNPs were undefined and it was unclear whether each SNP allele acted as an enhancer or repressor or was neutral. Instead, we relied on steric interference to reverse any effect of the SNP allele. Within each locus the various guide RNAs did not show equal effects on gene expression, reflecting that CRISPR-Cas9 can display considerable site-to-site variability in its activity, in some cases showing no activity. Nonetheless, for all three SNPs, CRISPRi yielded results that were concordant with the effects observed in
