Fig. 3

Effect of substitutions and temperature on bias for heterogeneously methylated templates. Two cell lines (KG1 and MDA-MB-231) were chosen that were respectively heterogeneously methylated for CDKN2B (panel a) and DAPK1 (panel b). The rows represent the different substitutions at the variable sites across the range of annealing temperatures used for each substitution. The numbers at the top are the CpG dinucleotides ordered across each amplicon. Only the CpGs that are not always methylated in the cell lines are shown. The final CpG in both assays has been omitted as the pyrosequencing measurement was called as inaccurate by the software. The annealing temperature increases along the y-axis by 2 °C for each step as indicated. Based on the homogeneous methylation data, where the lowest annealing temperature for the N-containing primers was the most accurate measurement of DNA methylation, the same conditions have been chosen as the standard for the measurement of DNA methylation in these tables (shaded in grey with the actual values measured). The values shown in the rest of the figure are deviations from the standard set of measurements (the raw data is shown in Additional file 1: Figure S1). The scale has been set so that minimal deviations from the standard (±5%) are in white. The colour scale indicating deviation from the standard is common for the two tables