*2.2. Genome Methylation Landscape of Prunus mume*

We used the BS-seq method, the "gold standard" of DNA methylation detection, to reveal the methylomes of six petal tissue groups from a single ornamental tree. In total, 40.0–50.9 million clean reads were generated, which corresponded to 11.6–14.6 Gb and greater than 41-fold coverage of the genome (estimated size = 280 Mb). To confirm the quality of the sequences, we also calculated QC20 (>97%) and QC30 (>92%) values, the bisulfite conversion rate (>99%), and GC content (21.37–21.54%). Approximately 59% of clean reads could be mapped to the reference genome, with a duplication rate of approximately 11.13–19.09% (Table S1). As revealed by sequencing coverage statistics, the maximum coverage was obtained at a sequencing depth of approximately 30× to 40× of the reference genome (Figure 2a), with each chromosome sequenced at a depth of around 24× to 34.2× (Figure S2). The coverage of cytosine sites, 8.8–10.95% of which were methylated, was reliably about 10.5× to 13.8×, with differing levels of mCG, mCHG, and mCHH sites (Figure S3, Table S2). *Int. J. Mol. Sci.* **2018**, *19*, x FOR PEER REVIEW 4 of 27 reads were generated, which corresponded to 11.6–14.6 Gb and greater than 41-fold coverage of the genome (estimated size = 280 Mb). To confirm the quality of the sequences, we also calculated QC20 (>97%) and QC30 (>92%) values, the bisulfite conversion rate (>99%), and GC content (21.37–21.54%). Approximately 59% of clean reads could be mapped to the reference genome, with a duplication rate of approximately 11.13–19.09% (Table S1). As revealed by sequencing coverage statistics, the maximum coverage was obtained at a sequencing depth of approximately 30× to 40× of the reference genome (Figure 2a), with each chromosome sequenced at a depth of around 24× to 34.2× (Figure S2). The coverage of cytosine sites, 8.8–10.95% of which were methylated, was reliably about 10.5× to 13.8×, with differing levels of mCG, mCHG, and mCHH sites (Figure S3, Table S2).

**Figure 2.** Genome sequencing coverage and methylation percentages and correlations between samples. WT1, WT2, and WT3 represent white petal tissue samples 1, 2, and 3, respectively. RT1, RT2, and RT3 represent red petal tissue samples 1, 2 and 3, respectively. (**a**) Distribution of the coverage of the six sequenced genomes. The *x*-axis represents sequencing depth, and the *y*-axis indicates the percentage of bases covered. (**b**) Percentage of methylated sites by sequence context (mCG, mCHG, and mCHH) relative to total methylated sites within each sample genome. (**c**) Pairwise correlations of methylation levels by sequence context (mCG, mCHG and mCHH) between **Figure 2.** Genome sequencing coverage and methylation percentages and correlations between samples. WT1, WT2, and WT3 represent white petal tissue samples 1, 2, and 3, respectively. RT1, RT2, and RT3 represent red petal tissue samples 1, 2 and 3, respectively. (**a**) Distribution of the coverage of the six sequenced genomes. The *x*-axis represents sequencing depth, and the *y*-axis indicates the percentage of bases covered. (**b**) Percentage of methylated sites by sequence context (mCG, mCHG, and mCHH) relative to total methylated sites within each sample genome. (**c**) Pairwise correlations of methylation levels by sequence context (mCG, mCHG and mCHH) between sequenced genomes.

To assess the influence of non-methylated cytosine on BS-seq library construction and control for methylcytosine sequencing preference [49], methylated loci mapped to unique sites in the reference genome were detected after M-bias assessment (Figure S4). The percentage of methylated cytosine to total cytosine, mC/C, was 13.72%, and the ratios of mCG/CG, mCHG/CHG and mCHH/CHH sites were 38.04%, 20.42% and 9.21%, respectively (Table 1). The relative proportion of mCG, mCHG, and mCHH sites throughout the genome was 28.28% (26.43–34.32%), 21.19% (20.24–24.46%), and 50.53% (41.42–53.26%), respectively (Figure 2b). Meanwhile, there existed a To assess the influence of non-methylated cytosine on BS-seq library construction and control for methylcytosine sequencing preference [49], methylated loci mapped to unique sites in the reference genome were detected after M-bias assessment (Figure S4). The percentage of methylated cytosine to total cytosine, mC/C, was 13.72%, and the ratios of mCG/CG, mCHG/CHG and mCHH/CHH sites were 38.04%, 20.42% and 9.21%, respectively (Table 1). The relative proportion of mCG, mCHG, and mCHH sites throughout the genome was 28.28% (26.43–34.32%), 21.19% (20.24–24.46%), and 50.53% (41.42–53.26%), respectively (Figure 2b). Meanwhile, there existed a tendency toward mCG

tendency toward mCG and mCHG sites with highly methylation levels; 70–100% were methylated with a large proportion. However, the mCHH context showed the opposite trend with a low and mCHG sites with highly methylation levels; 70–100% were methylated with a large proportion. However, the mCHH context showed the opposite trend with a low methylation level; 0–30% were methylated with a large proportion (Figure S5). An examination of logo plots exploring methylation preferences at sites and nearby regions revealed no significant variations in mC preference (Figure S6). Chromosome 3 had the highest number of methylated sites of any chromosome (Figure S7).


**Table 1.** Methylation of C contexts mapping to the reference genome. mC/C: methylated cytosine to total cytosine.

"WT1", "WT2" and "WT3" represent the white petal tissues; "RT1", "RT2" and "RT3" represent the red petal tissues.

Methylation levels of CG, CHG, and CHH sites were highly correlated (*R* <sup>2</sup> > 0.83) among different samples of the same petal color (Figure 2c), thus demonstrating that RT1–RT3 and WT1–WT3 could be analyzed as replicates of red petal tissue (RT) and WT samples, respectively. mCG, mCHG, and mCHH sites on chromosomes had similar distribution trends, with the most highly methylated regions in WT and RT corresponding to low-density genes (Figure 3a).
