In the relentless pursuit of oncology breakthroughs, pharmaceutical and biotech researchers are increasingly recognizing the limitations of single-omic datasets. Cancer is a multifactorial disease, shaped by a complex interplay of genetic, epigenetic, and transcriptomic alterations. To truly unravel these layers and develop more effective diagnostics and therapeutics, scientists need a multiomic perspective — one that can elucidate the full spectrum of biological changes driving tumor initiation and progression, as well as drug sensitivity or resistance.

This layered approach is becoming increasingly more achievable with the advent of new sequencing and spatial omic technologies that provide multiomic data on a single platform. Across research and development teams, there’s a growing consensus that extracting multiple data modalities from a single sample yields deeper insight, especially when working with precious or limited material. Yet, the conventional practice of using separate instruments and workflows for DNA, RNA, and epigenetic profiling introduces unnecessary complexity, increases turnaround time, and — most critically — risks losing the ability to compare data between platforms.

Data types: DNA, RNA, and methylation

DNA analysis has been a mainstay of drug discovery and development programs for decades. But not all sequencers provide the same quality or breadth of data, and it’s important to be aware of some limitations with common technologies. For example, the short-read sequencers that replaced Sanger instruments in most labs struggle to deconvolute structural variants or accurately represent repetitive sequence (1-3). By contrast, sequencers that are capable of reading DNA molecules of unrestricted read lengths can capture large variants in their entirety, and they are far better at analyzing long repetitive regions.

Meanwhile, DNA methylation has emerged as a critical data set to better understand cancer biology and diagnose disease. Aberrant methylation is increasingly recognized as an early event in tumorigenesis. Most sequencers cannot read methylation directly, so a bisulfite or enzymatic conversion step is needed. This adds to the cost and time to analysis, can damage precious samples or introduce errors, and requires that samples be split between two workflows — one for variant calling, and one for methylation sequencing. Some newer sequencers can read methylation without a conversion step, making it possible to analyze DNA and methylation together in the same sequencing reaction. Sequencing platforms capable of detecting a broad range of epigenetic markers will be instrumental in driving the discovery of the next generation of blockbuster cancer therapies.

Finally, RNA sequencing is an indispensable tool in cancer research, offering a real-time snapshot of gene expression that reveals how cancer cells activate, suppress, or rewire biological pathways. Most studies rely on cDNA rather than sequencing RNA directly — missing crucial features like RNA modifications that are lost during conversion (4). These modifications, part of the emerging field of epitranscriptomics, are increasingly linked to drug resistance and immune evasion. Direct RNA sequencing enables researchers to study RNA biology and the epitranscriptome comprehensively.

Multiomic discoveries

The best way to fully understand the value of a multiomic approach is to see how scientists are combining multiple layers of biological information to achieve new insights into cancer biology, classification, and treatment.

Continue reading below...

ExplainersWhy does immunotherapy work better for some cancers than others?

A powerful tool in modern oncology, immunotherapy doesn’t work the same for everyone. Researchers are exploring why and developing ways to improve its effectiveness.

Read More

At the University Hospital Heidelberg and Ghent University, clinical researchers developed a platform-agnostic workflow that incorporates DNA and methylation data to classify central nervous system tumors (5). By leveraging various methylation sequencing technologies, their approach achieved more than 99 percent accuracy in classifying methylation families and classes across nearly 80,000 samples. The study highlighted the unique value of nanopore sequencing, which can simultaneously deliver methylation and mutation data in real time — with preliminary results available in as little as 30 minutes. These rapid, multiomic insights can guide intraoperative decisions during tumor resection and later support more informed treatment planning through detailed molecular profiling.

Methylation was also a focus for scientists at the Dana-Farber Cancer Institute seeking a more accurate means of classifying acute leukemias by molecular subtype (6). This team developed a new framework based on nanopore technology to combine genome-wide DNA methylation data with machine learning. They built a reference of methylation profiles using more than 2,500 samples and controls from publicly available data sets including adult and pediatric leukemias and validated the resulting protocol with about 20 samples collected at their facility. Using the genomic and epigenomic data generated, they were able to generate accurate classifications for nearly 95 percent of cases. The authors were able to deliver classifications within two hours of receiving a sample, a vast improvement compared to the days-long — and often weeklong — wait time for conventional methods.

A similar effort from scientists at St. Jude Children’s Research Hospital and the University of North Carolina used chromosomal abnormalities paired with structural variant data to classify pediatric acute leukemia (7). By generating this data from a single analysis platform, they were able to define genomic subtypes in a matter of hours. “Whole genome nanopore sequencing with adaptive sampling has the potential to provide genomic classification of acute leukemia specimens with reduced cost and turnaround time compared to the current standard of care,” the team reported.

Finally, incorporating RNA data has allowed for considerable progress in cancer investigations. A group from the University of Washington used multiplexed adaptive sampling sequencing, aligned with the ability to sequence long or ultra long fragments of DNA, to analyze intronic variation in tumor suppressor genes, looking at both DNA and RNA results. (2) Separately, scientists at the H. Lee Moffitt Cancer Center and Research Institute and St. Jude Children’s Research Hospital integrated transcriptomic and proteomic data to identify cancer-specific transcript isoforms that could serve as novel therapeutic targets (8).

What’s next?

Adopting a single sequencing platform capable of generating multiple types of omic data is an important way to advance cancer-related drug discovery and development. Emerging technologies are beginning to integrate not only genomic and transcriptomic data sets but also proteomic insights, expanding their utility, scalability, and cost-effectiveness for researchers worldwide (9,10). Whatever the platform selected, though, it’s clear that multiomic investigations are here to stay — and that they already are shaping the future of cancer diagnosis, treatment, and precision medicine.

This article was contributed by Eleonora Juarez, the Global Segment Manager for oncology at Oxford Nanopore Technologies.

References

1. Cosenza, M.R. et al.Structural Variation in Cancer: Role, Prevalence, and Mechanisms. Annu Rev Genomics Hum Genet23, 123–152 (2022).
2. Gulsuner, S. et al.Long-read DNA and cDNA sequencing identify cancer-predisposing deep intronic variation in tumor-suppressor genes. Genome Res34, 1825–1831 (2024).
3. Erwin, G.S. et al.Recurrent repeat expansions in human cancer genomes.Nature613, 96–102 (2023).
4. Tang, Q. et al.RNA modifications in cancer. Br J Cancer129, 204–221 (2023).
5. Patel, A. et al.Prospective, multicenter validation of a platform for rapid molecular profiling of central nervous system tumors. Nat Med (2025).
6. Steinicke, T.L. et al.Rapid Epigenomic Classification of Acute Leukemia.Blood144 (Suppl 1), 273 (2024).
7. Geyer, J. et al.Real-time genomic characterization of pediatric acute leukemia using adaptive sampling. Leukemia (2025).
8. Shaw, T.I. et al.Multi-omics approach to identifying isoform variants as therapeutic targets in cancer patients.Front Oncol12, 1051487 (2022).
9. Lu, C. et al.Toward single-molecule protein sequencing using nanopores.Nat Biotechnol43, 312–322 (2025).
10. Motone, K. et al.Multi-pass, single-molecule nanopore reading of long protein strands. Nature633, 662–669 (2024).