Fig. 3
Multiplexed single-barcode system for targeted BS-Seq can accommodate a variety of strategies and can be extended to 478 individual indices. Total reads per index are outputed following each MiSeq run. a Using a smaller sample size, we are able to generate data where the percentage of reads allocated to each index is consistent and abundant. b Pooling indices with different ratios allows users to control the percentage of reads allocated to individual indices. c-d Our targeted BS-Seq protocol and the MiSeq sequencer can accommodate an extended number of indices, beyond 300 (c) and 478 (d), while still maintaining coverage greater than our threshold of 100×. a-d Y-axes correspond to the % of reads identified per indexed subject after sequencing. X-axes correspond to the number of indexed samples loaded onto the MiSeq. e Representative diagram of tiled primers on positive and negative strands of 11 loci interrogated by targeted BS-Seq, shown here for the CAMK2B locus as an illustration. These 11 loci were located on the following 7 genes: CAMK2B (1 locus), PAX6 (4), NDRG2 (2), LHX1 (1), ATP1A3 (1), TBR1 (1), and OTP (1). f Correlation of % modification values obtained using RRBS or our targeted BS-Seq approach (r 2 = 0.7922, p = 0.0002). The % of modification correspond to the mean values obtained across three biological replicates for both RRBS and targeted BS-Seq