Comparative Evaluation of DNA Methylation Platforms Reveals Context- and Region-specific Biases

DNA methylation is a key epigenetic modification essential for gene regulation, transposable element (TE) silencing, and genome stability. Genome-wide methylation profiling is now widely used in plant research, with whole-genome bisulfite sequencing (WGBS) being the most established method. However, bisulfite treatment can degrade DNA and introduce coverage bias. The recently developed enzymatic methyl-seq (EM-seq) minimizes DNA damage, while Oxford Nanopore Technologies (ONT) nanopore sequencing allows direct detection of methylated bases via ionic current shifts. Here, we applied WGBS, EM-seq, and ONT to identical genomic DNA extracted from an inbred line of Brassica rapa and compared methylation profiles across platforms. ONT exhibited the highest mapping rate and the most uniform genome coverage, especially in repetitive and GC-rich TE regions. WGBS showed biased coverage toward highly methylated regions, whereas EM-seq achieved the highest cytosine site depth despite the lowest mapping rate. Non-conversion analysis confirmed extremely low rates for both WGBS and EM-seq, while ONT showed a high false-positive rate for methylation calling, highlighting the need for improved signal-processing models. Although genome-wide methylation trends were broadly similar across platforms, ONT consistently reported lower CHH methylation levels in TE-rich heterochromatin. This likely reflects more accurate mapping in repetitive regions by ONT long reads, in contrast to WGBS’s overestimation in non-CG contexts. Pairwise correlation was high between WGBS and EM-seq, but substantially lower for ONT, particularly at CHH sites. Differentially methylated region (DMR) analysis further emphasized platform-specific biases, with the largest number of DMRs detected between WGBS and ONT. Together, our results highlight the distinct strengths and limitations of each platform and provide practical guidance for selecting optimal methylome profiling methods in future epigenomic studies of horticultural crops.