Credits
- Steven Ciaramaglia¹
- Jennifer Stone¹
- Aurelie Modat¹
- Robert Crawford¹
- Fabio Puddu¹
- Tom Charlesworth¹
- Robert Blanshard¹
- Annelie Johansson¹
- Luke Sarre¹
- Cillian Nolan¹
- Angela Simeone¹
- Robert Osborne¹
1 biomodal Ltd, Cambridge, UK
1. Introduction
DNA methylation plays a central role in gene regulation, but conventional epigenetic sequencing approaches either combine suppressive 5-methylcytosine (5mC) and activating 5-hydroxymethylcytosine (5hmC) into a single modified-cytosine signal or omit 5hmC entirely. This can obscure biologically meaningful changes because 5mC and 5hmC have distinct associations with regulatory activity and may change independently or in opposing directions relative to the healthy/control condition.
duet 6-base sequencing overcomes these limitations by independently resolving genetic sequence, 5mC and 5hmC from the same DNA molecules in a single assay. 6-base sequencing enables more complete detection and functional interpretation of differential methylation, whilst expanding the feature space available to study health and disease.
Fig. 1. The duet 6-base workflow¹ ligates a hairpin to DNA before a copy step preserves genetic information. 5mC and not 5hmC are copied onto the copy strand before protection and deamination steps. After conventional short read sequencing duet software resolves original and copied strands to produce 6-base data including A, T, G, C, suppressive 5mC and activating 5hmC.
2. 6-base sequencing resolves biologically distinct signals
5mC and 5hmC play complementary roles in gene regulation. Promoter-associated 5mC is generally linked to repression, whereas 5hmC is enriched at active enhancers and transcribed genes. Resolving both modifications independently therefore provides additional context for interpreting regulatory activity.
Fig. 2. Lung and liver tissue samples were prepared with duet and 6-base sequenced to >30x depth. Differential methylation analysis using modality XPLR software detected significant DMRs (q≤0.05 dashed coral line), revealing widespread repressive promoter 5mC hypomethylation and activating enhancer 5hmC hypermethylation in liver tissue compared to lung tissue.
3. 6-base differential methylation analysis reveals hidden changes in regulatory regions
Fig. 3A. Resolving 5mC and 5hmC independently identified 97% more promoter DMRs than modC analysis and revealed whether differential signals were associated with 5mC loss, 5hmC gain or changes in both modifications.
Fig. 3B. At enhancer regions, restricting analysis to 5mC-only missed approximately 33% of the DMRs detected through independent 5mC and 5hmC analysis, demonstrating the additional regulatory information contributed by 5hmC.
4. 6-base sequencing reveals hidden biology
Example liver biomarkers include crucial proteins involved in the removal of free plasma haemoglobin (HP – Haptoglobin) and triglyceride metabolism (APOA5 – Apolipoprotein A5). 6-base analysis reveals significant repressive and activating regulatory signal that would be hidden in modC or 5mC-only analyses.
Fig. 4A. At the HP promoter, liver tissue showed a significant loss of 5mC (coral) and opposing significant gain of 5hmC (green), consistent with activation. The corresponding modC signal is not significant as a consequence of summing 5mC and 5hmC, therefore masking a key regulatory change and contributing to the modC deficit observed in Figure 3A.
Fig. 4B. At the GH11F116733 enhancer, significant differential methylation gain (green) is observed for 5hmC, consistent with activation. The modC and 5mC signals are not significant, indicating the need for 6-base analysis to discover epigenetic changes in key enhancer regulatory regions, and contributing to the 5mC-only deficit observed in Figure 3B.
5. Epigenetic accuracy drives biomarker discovery
A high-confidence reference set of 9,368 promoter DMRs was identified from original methylation counts. Using the same underlying data, ten datasets were simulated for each of three 5mC accuracy profiles (Table 1). 5mC counts were retained according to sensitivity, while C and 5hmC counts were miscalled as 5mC according to 1-specificity. Promoter DMRs identified by modality XPLR (q < 0.05) were then compared with the reference set.
Table 1: 5mC accuracy profiles
Fig. 5A. Higher 5mC accuracy recovers more reference DMRs, increasing biomarker discovery. Lower-accuracy profiles missed biologically relevant liver biomarkers including KLF15 and MTTP.
Fig. 5B. Reduced 5mC accuracy increased false DMR discovery. Simulated datasets with lower sensitivity and specificity produced up to 2× more false-positive DMRs (y-axis) than duet 6-base mosaic.
6. Conclusions
- Independent analysis of 5mC and 5hmC reveals regulatory changes obscured by modC and 5mC-only approaches.
- 6-base analysis detected 97% more lung vs liver promoter DMRs than modC analysis, while 5mC-only analysis missed ~33% of DMRs driven by 5hmC changes.
- Higher CpG-level accuracy improves DMR recall and precision, supporting more complete and reliable biomarker discovery.
Together, these findings show that accurate, independent measurement of 5mC and 5hmC provides a more complete and biologically informative view of differential methylation.
7. References
- Füllgrabe J. et al. Simultaneous sequencing of genetic and epigenetic bases in DNA. Nat Biotechnol. 2023 Oct;41(10):1457-1464.
