Fig. 1: DNA methylation profiling and characterization of germinal center-derived B-cell lymphomas.

Using PGMRA on DNA methylation array data from germinal center-derived B-cell lymphomas (FL, DLBCL, FL-DLBCL), we identified 300 CpGs, which were subsequently organized into four modules (M1-M4) and seven methylation patterns (MPs) through k-means clustering. A Heatmap depicting DNA methylation levels of the 300 CpGs differentiating seven MPs (MP1-7). Sample features are annotated at the top of the heatmap. The mutation clusters were calculated for the entire dataset and DLBCL cases separately as described by Hübschmann et al. [8]. The TCC was calculated based on WGS data. Rows represent individual CpGs, and columns represent samples. CpG sites and samples are organized according to the k-mean clustering. B Radar plots illustrate the key defining features of each MP, including lymphoma classification, patient age, cell-of-origin (COO), BCL2 or BCL6 rearrangements, and mutational clusters identified by Hübschmann et al. C Boxplots display the distribution of biological age, Horvath’s epigenetic age, Ki-67 expression, and proliferation history (based on the epiCMIT package) across MPs. For statistical testing a pairwise Wilcoxon rank sum test with Bonferroni correction was applied (see Supplementary Tables S3 and S4). FL Follicular lymphoma, DLBCL Diffuse large B-cell lymphoma, ABC activated B-cell-like, GCB germinal center B-cell-like, TCC tumor cell content, WGS whole genome sequencing, r rearranged, n. a. not applicable.