Fig. 3: CHH methylation profiling of O. sativa centromere and transposable element regions. | Nature Communications

Fig. 3: CHH methylation profiling of O. sativa centromere and transposable element regions.

From: Accurate cross-species 5mC detection for Oxford Nanopore sequencing in plants with DeepPlant

Fig. 3

a Circos plot illustrating DeepPlant CHH methylation profiling in centromeric regions and 100 kb intervals upstream and downstream in O. sativa. From outer to inner: ideograms of centromere (center box) and neighboring regions (two terminal boxes); histograms of normalized sequencing coverage across 100 bp bins (gray, normalized against mean genomic coverage); histograms of CHH methylation frequencies (blue) across 100 bp bins. b Comparison of CHH motif coverage ratios across different genomic regions between BS-seq, DeepPlant, and Dorado profiling. To be noticed, the same nanopore dataset was used for DeepPlant and Dorado profiling, and the coverage difference between DeepPlant and Dorado derived from the distinct read filters they applied. DeepPlant applies three thresholds for screening high-quality alignments, including MAPQ ≥ 20, primary alignment length/read length ≥80%, and mapping identity ≥80% by default. Only CHH motifs with a minimum read coverage of 10 were regarded as quantified. TE transposable element, CDS non-TE protein-coding sequences. c Read coverage and CHH methylation frequencies in the centromeric regions of Chr1, comparing whole-genome BS-seq data, DeepPlant, and Dorado analysis on Nanopore data. d Boxplot illustrating CHH methylation frequencies on the forward (+) and reverse (−) strands in protein-coding and transposable element (TE) regions. The center line represents the median; each box shows the first and third quartiles; minima represents the larger between Q1−1.5×IQR and the minimum observed value; maxima represents the smaller between Q3 + 1.5×IQR and the maximum observed value. e Violin plot displaying strand-specific CHH methylation status across various TE types and non-TE protein-coding regions. The annotation of TEs and protein-coding regions was acquired from T2T-NIP40. Source data of (c, d) are provided in Zenodo [https://doi.org/10.5281/zenodo.15062213]. Source data of the other panels are provided as a Source Data file.

Back to article page