NEW YORK – Tagomics, a UK-based startup, has recently emerged from stealth with a new way to look at methylation, or lack thereof, in the human genome.
Last week, the firm announced a private funding round of £6.7 million ($8.5 million), led by Calculus Capital and joined by Illumina Ventures, IQ Capital, Agilent, Mercia Ventures, and the MEIF Proof of Concept and Early Stage Fund. It follows a £1.6 million pre-seed round led by IQ Capital and Start Codon, which also included grant funding from Innovate UK.
The funding will give Tagomics the chance to advance its workflows based around its Active-seq chemistry — a method for tagging and enriching unmethylated CpG sites for sequencing — and to develop its first products, which will target the pharmaceutical R&D market.
Active-seq is short for Azide Click Tagging for In Vitro Epigenomic sequencing, the proprietary chemistry developed by Tagomics Cofounder and CEO Jack Kennefick at the University of Birmingham, where he worked with Robert Neely, also a cofounder.
The method uses an engineered DNA methyltransferase to target unadorned CpG sites in the genome and label them via DNA alkylation with azide-terminated tags. An affinity probe then binds to the tags, and DNA fragments can be enriched with magnetic beads for sequencing. In a preprint posted to BioRxiv earlier this month, the Tagomics team claimed "Active-seq profile shows high enrichment efficiencies for unmodified CpG sites, near uniform enrichment efficiency across a range of CpG densities, and no significant off-target enrichment of DNA."
"Unmethylome markers are very important," Kennefick said. "They come from active regions like promoters, enhancers, partially methylated domains, and are often tissue specific." That could come in handy for Tagomics' ultimate goal of becoming a diagnostics company for cancer and other diseases where methylation has been demonstrated to play a role.
"Tagomics is a young company with cutting-edge technology," Tad Weems, managing director of Agilent venture investment, said in an email. The firm is part of Agilent's Early-Stage Partnerships program, which supports innovative startups with investment and consultation. "Their technology opens new possibilities for cancer diagnosis and treatment," he added.
Kennefick developed the technology underpinning Active-seq while a graduate student in Neely's lab and has matured and expanded it through a number of incubator programs. He applied for and won an ICURe fellowship from Innovate UK, which provided funding to do market research and explore commercial applications. "I flew around the world talking to senior executives at companies in a similar space to us to gauge whether there's appetite for our technology," he said. "There was, especially for liquid biopsy analyzing cfDNA."
From there, Kennefick joined the Royal Society of Edinburgh Enterprise Fellowship program and officially spun the company — along with the associated intellectual property — out of UB, with the help of the £1.6 million pre-seed financing. The firm then moved to Start Codon, a Cambridge-based accelerator.
Now, the firm has taken up residence as the first company in the new Illumina Ventures Labs, the successor to the Illumina Accelerator. It has seven employees and plans to increase headcount "fairly quickly," Kennefick said. He plans to pitch pharmaceutical companies on his firm's ability to combine analysis of unmethylated genomic regions with other genomic assays, such as cell-free DNA fragment analysis, gene panels, and low-pass whole-genome sequencing.
The "unmethylome," as Kennefick calls it, has the potential to provide for tissue-specific insights, he said, noting that it is correlated with active genomic regions. "It allows us to see which genes are switched on."
Kennefick noted that the preprint shows the ability to enrich the same tissue-specific markers presented as part of a DNA methylation atlas in normal cell types published in 2023 in Nature by an international team led by researchers from the Hebrew University of Jerusalem. That team wrote that "Loci uniquely unmethylated in an individual cell type often reside in transcriptional enhancers and contain DNA binding sites for tissue-specific transcriptional regulators."
As a transcription-suppressing epigenetic mark, DNA methylation, especially at CpG sites, has attracted a lot of attention in research and diagnostics, where it has been essential to clinical tests including Exact Sciences' Cologuard test for colorectal cancer, Guardant Health's Infinity assay, and the Grail Galleri test for pan-cancer early detection. Galleri assesses both methylated and unmethylated loci using a proprietary "high-yield" bisulfite conversion process on cfDNA, according to a Grail spokesperson.
While some long-read sequencing platforms can directly detect DNA methylation, other methods rely on a sample preparation workflow that converts unmethylated cytosines to uracil, which are in turn converted to thymine. This conversion can be done chemically, usually by treatment with bisulfite, or enzymatically.
These preparation methods can be expensive, Kennefick said, and bisulfite treatment may not fully convert unmethylated CpGs and can degrade DNA samples. Targeted methods using antibodies to pull out methylated sequences have their own flaws, and the unmethylated regions may provide more direct information on the biology of a sample, including tumor tissue of origin.
The "unmethylome" is also a smaller percentage of the whole genome, about 20 to 30 percent, compared to 70 to 80 percent that is methylated. "We can create an unmethylome profile from around 15 Gb of sequencing," Kennefick said, corresponding to about 5X coverage of the human genome. That pairs nicely with low-coverage whole-genome sequencing-based genotyping applications. It's also on par with the approximately 12 Gb to 15 Gb sequencing output per sample seen in the Circulating Cell-free Genome Atlas clinical validation study for Grail's multi-cancer early detection test.
Kennefick declined to disclose pricing and costs of the Active-seq method at this time, though the preprint states that it "provides a rich, biologically relevant perspective on the whole methylome at a fraction of the cost of base conversion chemistries."
Tagomics' workflows can use less than 10 ng cfDNA, Kennefick said. For comparison, Grail's clinical validation study for Galleri indicates that median cfDNA input was 15 ng and that for most cancer types, 3 ng is sufficient, a Grail spokesperson said in an email.
Tagomics is currently running proof-of-concept studies that Kennefick believes will demonstrate the technology's capabilities in different applications. The preprint includes an analysis of Active-seq profiles from tumor and adjacent normal tissues, analyzing differentially methylated regions.
"We're looking to partner with pharma to provide companies deeper biological insights into their clinical samples, to discover new drug targets, and to better stratify patients," he said. What form the company's first products will take is yet to be determined.
Ultimately, Kennefick wants to get into diagnostics, especially liquid biopsy assays. "We have promising data that shows the platform performs using cell-free DNA," he said. "And we've got promising results for early disease detection." Minimal residual disease is another application he mentioned, and the firm is interested in diseases beyond cancer, including cardiovascular and neurological diseases.
While the firm hasn't started selling its technology yet, Kennefick is confident in his market research. "I think there has been drive in the market for new epigenetic techs," he said. "For the longest time, innovation was stagnant. People were stuck with bisulfite and [methylation DNA immunoprecipitation sequencing], so they are welcoming the new technology."