DNA methylation is a highly studied epigenetic modification that regulates genome function and plays key roles in development and disease1. It is linked to a broad range of conditions, including inflammation, neurological disorders, and cancer. Some patterns of methylation are shared across cancer types, while others differ between subtypes – highlighting the value of studying DNA methylation to uncover novel biomarkers and gain insight into disease mechanisms, as well as drug mechanisms of action.
Due to its chemical stability, DNA methylation can be analyzed across a range of sample types, including fresh, frozen, and formalin-fixed paraffin-embedded (FFPE) tissues. Further, noninvasive approaches such as liquid biopsy are being increasingly used to measure methylation patterns, enabling new methods for diagnosis and patient monitoring. Emerging studies show that DNA methylation patterns in blood reflect disease-related changes in the brain, supporting their use as diagnostic biomarkers for neurodegenerative disorders2,3.
DNA methylation is also emerging as a promising therapeutic target, with multiple DNA methyltransferase inhibitors already approved for clinical use. Although several assays exist for mapping DNA methylation, better tools are needed to make these analyses more accessible and cost-effective for drug developers and translational researchers.
Tradeoffs in DNA Methylation Mapping: Cost, Coverage, and Resolution
Choosing the right method for DNA methylation profiling requires balancing several competing factors – cost, genome coverage, and resolution (Table 1). Each assay comes with its own tradeoffs, which not only impact data quality but also the feasibility of certain research applications.
Table 1: Comparison of DNA methylation sequencing technologies.
Whole-genome bisulfite sequencing (WGBS), the historical gold standard for genome-wide DNA methylation analysis, offers full coverage of the genome at base-pair resolution but comes at a high cost due to the need for high cell numbers and deep sequencing (>800 million reads per sample). Further, this assay uses a harsh bisulfite treatment that damages DNA and skews GC coverage, introducing biases that can overestimate DNA methylation levels. Non-destructive enzymatic approaches generate single-base resolution DNA methylation profiles and offer improved sensitivity at slightly lower sequencing depths4. However, both WGBS and enzymatic approaches remain inaccessible for many labs due to their high sequencing costs, intensive computational processing, and need for bioinformatics expertise5. Overall, these demands can make whole-genome sequencing approaches impractical for large-scale studies or experiments using limited or precious samples.
To reduce cost and complexity, researchers often turn to targeted approaches such as reduced representation bisulfite sequencing (RRBS), methylation arrays, or hybridization-based panels. These methods are more affordable but typically examine 3–15% of CpG sites and are biased toward CpG islands6,7. This limited scope may not provide the genome-wide coverage required for certain applications and can constrain the discovery of novel DNA methylation mechanisms.
Affinity-based techniques like methylated DNA immunoprecipitation sequencing (MeDIP-seq) are yet another strategy. This approach uses antibodies to enrich for methylated DNA, reducing sequencing requirements compared to whole-genome DNA methylation profiling approaches. However, MeDIP-seq is technically challenging and has several limitations, including the requirement for high cell numbers and a preference towards hypermethylated and low-GC content regions6,8. Because this approach uses immunoprecipitation and often relies on poor quality 5-methylcytosine (5mC) antibodies, MeDIP-seq suffers from poor reliability, resolution, and accuracy.
A Cost-Effective Approach to Mapping DNA Methylation
To address the limitations of existing DNA methylation profiling techniques, researchers are turning to EpiCypher’s CUTANA™ meCUT&RUN. A key advantage of this assay is its modular design, allowing researchers to choose between 1) direct sequencing of enriched fragments for efficient genome-wide mapping or 2) adding an enzymatic conversion step (NEBNext® EM-seq™ by New England Biolabs®) to achieve base-pair resolution.
Compared to targeted approaches like RRBS, microarrays, and hybridization panels, meCUT&RUN delivers broader and more uniform DNA methylome coverage with significantly lower sequencing requirements—achieving performance similar to EM-seq with only 20–50 million reads (Figure 1). Side-by-side comparisons show that meCUT&RUN identifies 80% of methylated CpGs (5mCs) captured by whole-genome EM-seq (Figure 2A), offering a highly sensitive strategy that balances cost and genome coverage.
Figure 1: meCUT&RUN provides a high-quality snapshot of the human DNA methylome in leukemia (K562) cells. meCUT&RUN paired with EM-seq for base-pair resolution readout (first track). RRBS (ENCODE) and EM-seq data (EpiCypher) are shown for comparison.
meCUT&RUN works by selectively enriching methylated DNA regions, avoiding harsh bisulfite-based conversion protocols that can degrade DNA. This approach maximizes yields while reducing cell input and sequencing requirements, making it especially useful for limited or precious samples. Compared to other enrichment-based approaches such as RRBS, meCUT&RUN detects more DNA methylation at enhancers, gene bodies, transcription start sites, and repetitive elements, highlighting its potential for innovative studies of the role of DNA methylation in gene regulation (Figure 2B).
Figure 2: meCUT&RUN generates DNA methylation profiles similar to whole genome EM-seq at greatly reduced sequencing depth. (A) meCUT&RUN recovers 80% of methylated CpGs (5mCs) compared to whole-genome EM-seq in K562 cells, with just 20-30 M unique reads. meCUT&RUN was paired with EM-seq for base-pair resolution 5mC analysis (meCUT&RUN-EM) and downsampled from 50 to 3 M uniquely aligned reads. The number of methylated CpG positions was calculated for each downsampled dataset and normalized to the number of 5mC position detected by EM-seq. (B) Methylated CpGs detected by each method were assigned to respective genomic features; total counts were normalized to EM-seq data.
Unlocking the Full Potential of DNA Methylation Research
The trade-offs between sequencing cost, depth, and coverage have long hindered DNA methylation studies. Newer approaches like meCUT&RUN offer a more flexible solution that supports large-scale pharmaceutical and clinical studies, without sacrificing coverage or breaking the budget. By making DNA methylation analysis more accessible, this technology paves the way for the development of novel diagnostics and targeted therapies, bridging the gap between basic research and clinical application.
References
1 Mattei, A. L., Bailly, N. & Meissner, A. DNA methylation: a historical perspective. Trends in Genetics 38 (2022/07/01). https://doi.org/10.1016/j.tig.2022.03.010
2 Wei, X., Zhang, L. & Zeng, Y. DNA methylation in Alzheimer's disease: In brain and peripheral blood. Mech Ageing Dev 191, 111319 (2020). https://doi.org/10.1016/j.mad.2020.111319
3 Miranda-Morales, E. et al. Implications of DNA Methylation in Parkinson's Disease. Front Mol Neurosci 10, 225 (2017). https://doi.org/10.3389/fnmol.2017.00225
4 Wang, T., Loo, C. E., & Kohli, R. M. Enzymatic approaches for profiling cytosine methylation and hydroxymethylation. Molecular metabolism, 57, 101314 (2022). https://doi.org/10.1016/j.molmet.2021.101314
5 Gong, T. et al. Analysis and Performance Assessment of the Whole Genome Bisulfite Sequencing Data Workflow: Currently Available Tools and a Practical Guide to Advance DNA Methylation Studies. Small Methods 6, e2101251 (2022). https://doi.org/10.1002/smtd.202101251
6 Beck, D., Ben Maamar, M. & Skinner, M. K. Genome-wide CpG density and DNA methylation analysis method (MeDIP, RRBS, and WGBS) comparisons. Epigenetics17, 518-530 (2022). https://doi.org/10.1080/15592294.2021.1924970
7 Sun, Z., Cunningham, J., Slager, S. & Kocher, J. P. Base resolution methylome profiling: considerations in platform selection, data preprocessing and analysis. Epigenomics 7, 813-828 (2015). https://doi.org/10.2217/epi.15.21
8 Lentini, A. et al. A reassessment of DNA-immunoprecipitation-based genomic profiling. Nat Methods 15, 499-504 (2018). https://doi.org/10.1038/s41592-018-0038-7