with Mecp2e1 (r = -0.9, P <0.05), while CpG2 showed a significant negative correlation with Mecp2e2 (r = -0.9, P <0.01) (Figure 7B, c). Further confirming the potential role of these two CpG sites within R3 in Mecp2 isoform-specific expression, a similar correlation (CpG1: Mecp2e1, r = -0.85, P <0.05; CpG2: Mecp2e2, r = -0.87, P <0.05) was observed at D8 (Figure 7C, c). The studied intron 1 regions seemed to have preferential correlation with individual isoforms. For instance, at D2, the only CpG site within R5 showed negative, significant correlation with Mecp2e1 (r > -0.8, P <0.05) (Figure 7B, e). Interestingly, at D8 intron 1 R4 and R6 showed positive, significant correlation with the Mecp2e2 isoform (r >0.8, P <0.05) (Figure 7C, d-f). The observed correlations for all REs are represented in Figure 7B-C and are summarized in Figure 8.
