NEW YORK – UK DNA sequencing technology firm Cambridge Epigenetix has rebranded itself as Biomodal, launching a product this week that promises to capture genetic and epigenetic information in DNA with a single sequencing workflow.
Established in 2012, Cambridge Epigenetix was born out of cofounder Shankar Balasubramanian’s lab at the University of Cambridge to develop and commercialize an oxidative bisulfite sequencing (oxBS-seq) technology that can quantitatively analyze 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) in genomic DNA.
“The company went through some iterations of that technology,” said CEO Peter Fromen, who joined the firm from Pacific Biosciences in June of last year. Eventually, Cambridge Epigenetix devised an approach that can detect methylated cytosine as well as the four DNA bases in the same sequencing read and decided to commercialize that workflow, he said, and raised $88 million in Series D financing in November 2021.
Dubbed duet multiomics solution +modC, the renamed company’s first product is a sequencing sample prep kit to simultaneously detect canonical DNA bases (A, C, G, T) and modified cytosines in a DNA sample. In addition, the firm said it is releasing a “plug-and-play” software that uses open-source and proprietary algorithms.
The firm's new name, Biomodal, underscores its technology’s capability to layer different modalities of biological insights with genetics information, Fromen noted. “The name Cambridge Epigenetics is a bit backward-looking and provincial,” he said. “We felt that it was critical to rebrand the company to highlight and be specific about where we're going.”
Biomodal’s product launch was bolstered by a recent Nature Biotechnology paper, led by Balasubramanian and company scientists, which offered a closer look at the inner workings of the company’s core technology.
“DNA epigenetics has fundamental information on dynamic biology,” said Balasubramanian, who also serves as Biomodal’s scientific adviser. “To me, the concept of capturing dynamic changes to DNA that reflect the dynamic shifts in biology — whether it's developmental biology, or the onset of disease, or adaptation to the environment — is all very important.”
Previously, researchers have mostly been using base conversion methods, such as bisulfite sequencing, to help distinguish unmodified cytosines from their methylated variants.
However, Balasubramanian said these methods typically come with pitfalls. For instance, with bisulfite-treated DNA, “it's very hard to accurately get genetic information," he said. In addition to degrading DNA molecules, converting unmethylated cytosines to uracils means one cannot unambiguously capture common C-to-T mutations in the genome, he noted.
Furthermore, existing methods typically don't simultaneously profile the full genetic and epigenetic signatures of a sample, resulting in a lag between the two layers of information. “There is an interaction, biologically speaking, between epigenetics and genetics,” Balasubramanian said. “You need to really capture all of those dimensions of information at the read level, and you don't get that if you do parallel experiments.”
To that end, Balasubramanian’s team devised a so-called "five-letter seq" workflow that uses a two-base coding strategy to account for each base as well as modified cytosine.
A DNA molecule first undergoes a series of proprietary enzyme transformations to produce a covalently linked complementary strand. During that process, the methylation status of cytosines is relayed to the copy in the format of a base code.
“At the end of the day, when you feed the sample into the sequencer, the sequencer will read the copy strand and the sample strand,” Balasubramanian explained. “For each base that is read, there's a cognate base, so you get combinations of two bases.”
The sequencing data is then processed using software that resolves paired bases from the sample and copy strand into one of five states: A, C, G, T, or modified C.
Using the same principle, Balasubramanian’s team also demonstrated the feasibility of six-letter sequencing, which can determine the four bases plus both 5mC and 5hmC at the same time.
Compared with previous methods, Balasubramanian said that five-letter seq, which employs an all-enzyme workflow with “extremely high conversion efficiencies,” can minimize damage to the DNA molecules. Additionally, he said the approach is relatively easy to carry out and results in “a fairly standard sequencing run” after the initial enzymatic workflow.
Notably, during data analysis, the resolution algorithm can also account for errors produced by unexpected base pairings between the original strand and its copy. The phased readout provides an “added bonus” to the technology by suppressing PCR or sequencing errors, leading to improved accuracy, Balasubramanian noted.
In their study, the researchers applied five-letter seq to human genomic DNA and cell-free DNA from a blood sample of a stage III colon cancer patient, using as little as 2 ng of input DNA. Overall, Balasubramanian said the team was “very pleased” with how well the technology performed. “For the kinds of diagnostic applications such as liquid biopsy analysis, you need to work with low nanograms of input DNA,” he said.
The publication “really represents the foundation of a platform that we define as our core technology as a company,” said Fromen, and the company owns intellectual properties pertaining to the enzymology of the five-letter seq technique.
The newly launched five-letter seq product is a pre-sequencing kit that comes with eight or 24 reactions to generate sequencing-ready libraries with ligated Illumina adaptors, Fromen said, and Biomodal is making its proprietary analysis pipeline available. According to the company, the DNA input for the kit can be as low as 10 ng.
While the initial five-letter seq product is optimized for Illumina sequencing, Fromen said the company’s technology is “absolutely platform agnostic” and requires no capital investment that a customer has to make.
In terms of turnaround time, Fromen said the workflow currently takes roughly two days to complete, but the company thinks there is still “plenty of headroom to optimize.”
Biomodal did not disclose the price of its kit but said it “will be a premium solution.”
Fromen said there are four primary application markets the company is eyeing. These include oncology, such as tumor liquid biopsy and early cancer detection; neurodegenerative disorders research; cellular development and reprogramming, particularly in the context of aging; and prenatal screening.
While Biomodal does not have plans to develop diagnostic products, Fromen said it will “partner and support those who are seeking to do so.”
As single-molecule sequencing technologies continue to evolve, companies such as PacBio and Oxford Nanopore Technologies are also promoting their ability to simultaneously profile genetics and epigenetics signals, and it remains to be seen how Biomodal will compete with these approaches.
Fromen believes that one competitive advantage of the five-letter seq technique is that it can work with low-input samples. Single-molecule sequencing technologies “require a significant amount of high molecular weight DNA,” he said, since they are incompatible with PCR amplification of the starting material. Liquid biopsies in particular, with nanogram levels of input DNA, are "just not an addressable market for the single-molecule technologies,” he said.
However, some researchers have successfully been using single-molecule platforms to profile cancer-specific epigenetic changes from liquid biopsies.
Additionally, since both PacBio and Oxford Nanopore rely on analog signals to call methylated bases, a Biomodal spokesperson argued these technologies are “subject to significant noise and therefore potential error.” In addition, Biomodal said it gives owners of short-read sequencing platforms the opportunity to perform joint epigenetic and genetic analysis “without having to purchase new, expensive capital equipment.”
However, the company acknowledged that given there are “few relevant publicly available datasets” available, head-to-head comparisons between its technology and the single-molecule sequencing approaches have been “minimal.”
Along with its product launch, Biomodal is growing. The firm has already more than doubled its headcount since last year and will continue to expand its commercial team to support its business ventures.
In addition to five-letter seq, the firm will also work to commercialize six-letter seq, Fromen said. Beyond that, it aspires to incorporate many other epigenetic modifications of DNA into its assay.
“There are numerous modalities of epigenetics that you can start interrogating and layering in conjunction with the genetics,” Fromen said. “That's what that construct of the assay allows us to do.”