Products
Explore our advanced sequencing products for turning complex molecular signals into high-confidence readouts across clinical testing, RNA quality control, and translational methylation programs.

Authors: The Renew Scientific Team
Can DNA methylation serve as a biomarker for autoimmune disease?
DNA methylation is a promising but not yet validated source of biomarkers for autoimmune disease. It can reflect disease activity and treatment response that genetic risk alone cannot fully explain,² but most candidates have not yet been reproduced across independent cohorts. Moving a candidate from discovery to validation takes broad methylome profiling first, then targeted testing in larger, independent cohorts.
The need is real. Among the estimated 15 million Americans diagnosed with at least one autoimmune disease,¹ a shared diagnosis does not always mean a shared disease experience. Symptoms, organ involvement, and treatment response vary widely, which makes it hard to find biomarkers that reliably track an individual patient’s disease activity and progression.
Why is DNA methylation relevant to autoimmune disease?
DNA methylation is an epigenetic mechanism involved in gene regulation, with patterns that vary by cell type and immune activity2,3. Environmental exposures, hormones, infections, and other non-genetic factors have also been associated with autoimmune disease development. Epigenetic mechanisms may explain how some of these factors interact with immune biology2.
In multiple sclerosis (MS), Xavier and colleagues found that DNA methylation signatures occurred independently of known genetic risk and were attributed primarily to B cells and monocytes4. The findings show how strongly a sample’s cellular composition can influence methylation results.
Methylation patterns may change with disease activity. In a longitudinal study of 59 patients experiencing an active systemic lupus erythematosus (SLE) flare, Horton and colleagues measured methylation during the flare and approximately three months later. Changes at 1,953 CpG sites differed between patients who achieved remission and those who did not. The researchers identified three patient clusters based on methylation changes, including one in which all 12 patients achieved complete remission at follow-up5.
The results suggest that methylation may help reflect biological differences among patients with the same disease, although the study was small⁵.
For drug developers, this points to a practical application. Autoimmune trials often enroll patients who share a diagnosis but respond differently to the same treatment. Methylation profiles that track disease activity or predict response could support patient stratification and treatment-response monitoring, although any such marker still requires validation in independent cohorts5,13.
Can a blood test show which tissues autoimmune disease is affecting?
Cell-free DNA (cfDNA) can be used to discover methylation biomarkers. As cells die, they release DNA fragments into the bloodstream. These fragments can retain methylation patterns associated with their cell or tissue of origin. A blood sample may therefore provide information about activity in tissues that are difficult or impractical to sample directly, with potential applications in disease characterization and monitoring6,7.
Lehmann-Werman et al. demonstrated this approach by using tissue-specific methylation patterns in circulating DNA to identify DNA originating from pancreatic β-cells involved in type 1 diabetes and oligodendrocytes involved in relapsing MS⁷.
More recently, Liu et al. used whole-genome bisulfite sequencing to profile plasma cfDNA methylation in patients with lupus nephritis and developed a methylation-based model that was subsequently evaluated in independent SLE datasets8.
These approaches present technical challenges for study design and interpretation. cfDNA is present at low abundance and is highly fragmented, which can make analysis difficult. Cell-of-origin analysis depends on accurate reference methylomes, and variation in the tissues and cell types contributing DNA to a sample can complicate interpretation6,8.
For now, cfDNA methylation in autoimmune disease is an emerging area of research rather than an established clinical tool6,8.
Which methylation profiling method fits a biomarker study?
No single DNA methylation profiling method fits every biomarker study. Researchers must consider whether the study focuses on broad methylation discovery, specific genomic regions, or additional sequence and haplotype information, along with practical factors such as DNA input, throughput, and computational requirements. Arrays, bisulfite-based sequencing, enzymatic methods, and native long-read sequencing each have advantages and limitations9,10.
Methylation arrays allow researchers to assess predefined CpG sites across many samples. However, the widely used MethylationEPIC array measures roughly 850,000 of the approximately 28 million CpG sites in the human genome, so signals outside that fixed content go undetected9,10.
Bisulfite sequencing has historically been used for whole-genome methylation analysis, but the treatment can alter or damage DNA9,11. Native long-read technologies avoid bisulfite conversion, allowing direct detection of base modifications in native DNA. Longer reads can also preserve genomic context for variant phasing and haplotype-resolved analysis10,11.
Benchmarking studies highlight these trade-offs. The SEQC2 Epigenomics Quality Control study compared whole-genome bisulfite sequencing, enzymatic deamination, targeted sequencing, nanopore sequencing, and methylation arrays, identifying differences in mapping efficiency, CpG capture, coverage, and platform performance9.
Brooks et al. later compared nanopore sequencing with the MethylationEPIC array and, separately, with enzymatic methyl-sequencing (EM-Seq). In four blood samples, nanopore sequencing and the MethylationEPIC array showed correlations of at least 0.94 across approximately 850,000 shared CpG sites. The methods differed in coverage, DNA requirements, cost, and computational demands10. That concordance shows nanopore calls are accurate where the two methods overlap. The larger difference is reach: native sequencing reads methylation across the genome, not only at predefined array sites10.
What does native long-read sequencing add to methylation analysis?
Oxford Nanopore sequencing measures changes in electrical current as DNA passes through a nanopore. Modified bases alter the signal, allowing methylation to be detected directly from native DNA9.
Nanopore sequencing has also been used to detect both 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC), including strand-specific modification patterns12.
Long genomic DNA fragments allow methylation calls to be considered together with sequence variation and haplotype context. Researchers can examine whether an epigenetic pattern is associated with a particular allele or genomic region rather than viewing methylation sites in isolation10,11.
cfDNA is different. Circulating DNA is already highly fragmented, so much of the long-range context available from intact genomic DNA is lost. For cfDNA studies, native sequencing offers the ability to detect methylation directly without chemical or enzymatic conversion6,11. This matters most when input is scarce. Bisulfite conversion can lose much of the starting template. In a comparison of 12 commercial kits, mean recovery of cfDNA-like DNA after conversion ranged from 9% to 32%.14 Native sequencing avoids that conversion step entirely.
What does it take to validate a methylation biomarker?
A statistically significant methylation signature is not automatically a biomarker. Cell composition, sample type, profiling method, ethnicity, and other population differences can all affect the signals detected in a study4,13. Candidate markers therefore need to be tested in independent cohorts and under conditions that account for these sources of variation.
SLE research shows how difficult that can be. A systematic review of 44 case-control studies involving 3,471 patients identified several recurring candidates, including IFI44L, FOXP3, and MX1. The authors found conflicting results across studies and cited sample type, methylation profiling methodology, ethnicity, and shared methylation patterns across autoimmune diseases as possible confounders. Based on the available evidence, they concluded that DNA methylation markers could not yet be considered reliable diagnostic or surrogate disease-activity biomarkers in SLE13.
Discovery and validation may also call for different methods. Whole-methylome profiling enables researchers to examine methylation patterns broadly and identify candidate regions associated with disease. Once promising candidates have been identified, targeted approaches can be used to test those regions in larger or independent cohorts without repeating genome-wide profiling for every sample9,11.
How does Renew support methylation biomarker studies from discovery to validation?
Renew generates methylation data that conventional assays structurally cannot: native methylation, read together with sequence and haplotype context from the same molecules. Renew offers whole-methylome, targeted, and cfDNA methylation approaches for biomarker research. Most programs start with a focused discovery pilot. Direct Whole Methylome Sequencing (dWMS) provides genome-wide methylation profiling from native genomic DNA, with sequence and haplotype context retained from the same reads. Direct Targeted Methylation Sequencing (dTMS) then tests the strongest candidates across larger, independent cohorts without repeating genome-wide profiling for every sample.
For blood-based studies, Renew supports cfDNA methylation and cell-of-origin analysis. Bioinformatics, assay development, and assay validation are available for studies expanding into larger cohorts or moving toward translational applications. This work is performed within a CAP-accredited, CLIA-certified laboratory environment.
Learn more about Renew’s Methylation Solutions
Scope a methylation discovery pilot with the Renew team