Fig. 2

The distinct clustering patterns of disease-specific DHMRs. A The t-SNE plot for the merged DHMRs identified from the analyzed cohort. Three distinct clusters were observed, reflecting distinct disease status. B Kaplan–Meier survival curves for patients with high and low 5hmC levels at selected genomic regions. The high and low 5hmC were separated by the median value of 5hmC. Boxplot: Bounds of the box span from 25 to 75% percentile, the center line within each box represents the median. Whiskers represent median ± 1.5 times interquartile range. C The clustering analysis of DHMRs based on TET2 mutation status in healthy controls and patients with myeloid neoplasms (with WT or mutant TET2). D The immunophenotypic features, including platelet (PLT), blasts, and monocytes (mono) counting in patients with WT TET2 and mutant TET2. E Venn diagram of DHMRs identified between the comparison of healthy donors vs patients with WT TET2 or healthy donors vs patients with mutant TET2. F GREAT analysis of DHMRs identified from the comparison shown in (E)