Decoding Biology at the Single Cell Level: Why DNA Methylation Is the Next Frontier

Single cell analysis is transforming biology - but what lies beyond RNA? Single cell analysis has emerged as a revolutionary technology, fundamentally changing our understanding across diverse areas in biology. While most breakthrough discoveries have been powered by single cell RNA sequencing, new frontiers are opening as additional single cell -omics methods become more accessible and robust. 

One particularly promising yet underutilized approach is single cell DNA methylation. This powerful epigenetic profiling method offers unique insights impossible to obtain through transcriptomics alone. Let's explore what makes single cell methylation so valuable and examine some research examples that demonstrate its unique capabilities. 

First thing’s first – is DNA methylation only measuring transcriptional (in)activity? 

In a word, no! Although DNA methylation is an important epigenetic marker generally associated with transcriptional repression, when profiled at the single cell level, it can also be a powerful tool for:

Classification with unprecedented resolution – DNA methylation can identify tumor subtypes with single cell precision 

Lineage tracing – DNA methylation acts as a "natural" heritable barcode for cell lineage tracing without artificial manipulation 

Genomic structural analysis – DNA methylation can be used to measuring copy number variation or alteration (CNA) and identifying subclones within tumors 

Biomarker discovery for non-invasive testing – DNA methylation can reveal cell type- or cell state-specific biomarkers for clinical use 

One single cell methylation experiment can lead to multiple levels of inquiry, increasing discovery power. Let’s take a closer look at a few examples where single cell methylation was instrumental

A mysterious case of unknown tumor identity  

To study pediatric glioma tumors, Dr. David Rincón Fernandez Pacheco and colleagues at Cedars Sinai use mouse models containing commonly mutated driver genes. Using single cell RNA-seq, they characterized transcriptional diversity and cell types in each tumor subtype. However, the cells from different subtypes clustered together when performing UMAP analysis, making it difficult to distinguish glioma tumors by scRNA-seq. How could they begin to examine what makes these tumors different from each other? 

In addition to single cell RNA-seq, they also profiled the different tumor subtype models with single cell ATAC-seq and single cell methylation. Similarly to scRNA-seq, scATAC-seq wasn’t able to discern the different subtypes. Only single cell DNA methylation was able to classify the tumor subtypes from the three mouse models – allowing the investigators to start investigating critical differences that may have clinical impact. Learn more in a webinar Dr. Rincon gave here.   

All-natural barcodes – no artificial sequences needed  

You’ve heard of mutations, but have you heard of epimutations? These changes in DNA methylation accumulate more rapidly than DNA mutations but most are stochastic passenger events that do not impact gene regulation. These heritable changes can be used as a natural barcode to trace cell lineage at high resolution.  

For example, Dr. Dan Landau, Dr. Mario Suvà and colleagues at New York Genome Center and Massachusetts General Hospital used single cell DNA methylation to reconstruct cell lineages in two forms of glioma (Chaligne et al 2021). While the less aggressive form, IDH-mutant glioma, cell states (from scRNA-seq) neatly fit within the DNA methylation-based lineage tree, the more aggressive IDH-wildtype glioblastoma (GBM), multiple different cell states were found within individual DNA methylation-based clades, suggesting higher cell state plasticity.  Learn more about reconstructing cancer cell lineages in this Bioinformatics Working Group with the co-first author of the study, Dr. Federico Gaiti. 

Beyond cancer research, these epimutations serve as "molecular clocks" in aging-related research, allowing scientists to track cellular age and history with unprecedented precision. The higher rate of epimutation accumulation provides finer temporal resolution than genetic mutations alone could offer.  

Building more powerful atlases for non-invasive testing  

Non-invasive clinical specimens such as urine or blood contain short fragments of methylated DNA shed from cells, including potentially disease-altered cells. Profiling cell-free DNA (cfDNA) from these samples is a promising method for detection and monitoring of several diseases including cancer and brain disorders (Tian et al 2023, Cisneros-Villanueva et al 2022), as DNA methylation can serve as an accurate biomarker to identify pathogenic cells or states.  

Non-invasive tests rely on reference datasets from healthy and diseased tissue. Reference datasets are typically generated using bulk analysis, which has limited power to resolve cell types and states. Ecker and colleagues demonstrate that with single cell DNA methylation and machine learning, they can reliably identify brain cell types using  ~200 methylated CpG sites. This DNA biomarker signature could potentially be used as part of a non-invasive test for detecting pathogenic brain cell types, transform the diagnosis of brain disorders (Tian et al 2023). 

Technology Making It Possible

You might be wondering – if single cell methylation is so powerful, why haven't we been doing it all along? The answer is simple: it's technically challenging. 

Until recently, analyzing DNA methylation in individual cells required physically isolating individual cells or nuclei into wells and labor-intensive (and expensive) workflows. However, new technologies like Scale Bio's Single Cell Methylation kit have changed that by leveraging massively parallelized barcoding technology. This means researchers can now analyze tens of thousands of cells simultaneously while detecting hundreds of thousands of methylation sites per cell.  

What's Next for DNA Methylation Research? As these single cell technologies become more accessible, we're entering an exciting new era for epigenetics research. We're moving from pixelated, low-resolution views to crystal-clear images of DNA methylation states in every cell. 

The bottom line? Single cell methylation isn't just an incremental improvement over bulk methods – it's revealing an entirely new layer of biology that's been hiding in plain sight all along. And that's something worth getting excited about. 

Want to learn more?

• Download our eBook 
• Read about our Single Cell Methylation Product  

References

Chaligne R, Gaiti F, Silverbush D, et al. 2021. Epigenetic encoding, heritability and plasticity of glioma transcriptional cells states. Nat Biotech doi: 10.1038/s41588-021-00927-7 

Cisneros-Villanueva M, Hidalgo- Pérez L, Rios-Romero M, et al. 2022. Br J Cancer. doi: 10.1038/s41416-021-01696-0  

Shao X, Lv N, Liao J, et al. 2019. Copy number variation is highly correlated with differential gene expression: a pan-cancer study. BMC Med Genet doi: 10.1186/s12881-019-0909-5  

Tian W, Zhou J, Bartlett A, et al. 2023 Single-cell DNA methylation and 3D genome architecture in the human brain. Science doi: 10.1126/science.adf5357