How do our cells organise?
Researchers are uncovering a new layer of epigenetic information inside our cells. Decoding its language could help combat cancer and neurodegenerative diseases.
You’re probably familiar with the ‘blueprint’ metaphor for our genetic code: that within our DNA lies a set of instructions for making our bodies.
Epigenetics tells us that the real world is more complicated. In recent years, scientists have discovered a set of chemical modifications sitting on our DNA that regulate gene activity. These modifications influence how and when genes are activated, promoting some parts of our code while suppressing others.
The developing picture makes our DNA more like a cookbook than a blueprint, where epigenetic processes mix and match genetic ingredients to produce a specific dish (us!).
Epigenetic processes don’t change the genetic code itself. Instead, by taking cues from our environment, they fold experience into our development, making us more flexible in response to our surroundings.
Using the epigenetic chef, nature can cook up many phenotypes (body traits) from the same genotype (genetic code). It allows cells that inherit the same genetic information to do different things with the code – by making our body’s various cell types, for example.
Now, Cambridge researchers in the group of Professor Sir Shankar Balasubramanian FRS have found what appears to be another layer of information in mammalian DNA.
The group developed a new sequencing method, called SCoTCH-seq, that can map different forms of cytosine (the ‘C’ in ‘ACGT’, the 4 bases found in a DNA molecule).
Their technology can read 3 kinds of cytosine on both strands of DNA, allowing them to map these features in stem cells for the first time.
Two of cytosine’s modifications – methylation and hydroxymethylation (hmC) – are central to epigenetics. HmC in particular remains a mysterious player; although seemingly essential to our development, researchers contest its role and function.
Dr Jack Hardwick, research lead and former group member, thinks that hmC contains vital information that instructs cells on how to use their genomes.
If he’s right, untangling this information could let us better track the development of cancers, and identify potential treatments for neurodegenerative diseases.
“We could use this information to predict gene expression,” says Jack. “It could be important for diagnostics and assessing disease progression.”
This discovery was made possible by a series of Cambridge breakthroughs, from immortalising embryonic stem cells to pioneering sequencing technologies. Researchers are now piecing together how epigenetic information is encoded across the DNA double helix to reveal hidden chapters in the story of ourselves. Here’s how they did it.
“It’s like finding out how letters combine, in a language we are still trying to learn.”
Dr Jack Hardwick, former Postdoctoral Research Associate and Leverhulme Trust Early Career Fellow in the Yusuf Hamied Department of Chemistry
The road less travelled
As a Postdoctoral Research Associate and then Leverhulme Trust Early Career Fellow in the Yusuf Hamied Department of Chemistry, Jack was in the Balasubramanian Group for 5 years. He’s now heading up the Laboratory of Neuroepigenetics at the University of Bristol.
Ironically, Jack’s restlessness at school meant he was banned from his last chemistry lesson. Leaving school at 16, young Jack wanted to pursue a career in music. For a while he soundtracked films and advertisements, while doing odd jobs like cleaning toilets in hospitals.
Jack later found within himself a desire to teach, and realised he couldn’t do so without a degree. Getting back into education was a challenge.
He did part of a Maths A-level in a local school, alongside people who looked half his age. But having succeeded, he catapulted into the Science Foundation Year at the University of Southampton.
His newfound passion led him to a PhD at the University of Oxford, where he focused on chemically modified DNA.
Jack says, “That’s when I first became interested in these epigenetic combinations, and how they could affect the structure of DNA.”
When Jack started at Cambridge in late 2019, he joined the Balasubramanian group, which has been working on DNA modifications for over 10 years. The group’s leader, Professor Sir Shankar Balasubramanian, pointed him in the direction of new sequencing techniques to go at the problem.
Shankar is a giant in the field. He won the Millenium Technology Prize with Professor David Klenerman in 2021, for developing revolutionary DNA sequencing techniques. His Solexa-Illumina technology underlies the majority of DNA sequencing conducted around the world.
“This new technique is the culmination of a conversation Shankar and I had,” Jack says.
The new sequencing method was developed in collaboration with biomodal, a company launched by Shankar and Bobby Yerramilli‑Rao in 2013.
Biomodal developed the initial technology to detect both genetic and epigenetic information in a single sample. Their method could distinguish between the three most common forms of cytosine in single-stranded DNA.
The modifications mC and hmC can form different combinations across the DNA double helix. SCoTCH-seq can read these different double-stranded combinations, revealing previously unknown information.
The modifications mC and hmC can form different combinations across the DNA double helix. SCoTCH-seq can read these different double-stranded combinations, revealing previously unknown information.
Jack adapted this technology to read both DNA strands simultaneously. This enables researchers to see new combinations between the different kinds of cytosine that form across the double helix, and measure where these occur throughout the entire genome.
“It’s like finding out how letters combine, in a language we are still trying to learn,” Jack says. “These double-stranded combinations form another layer of information in DNA, which we can now read.”
What might this new language be telling us?
How do our cells organise?
Because hmC occurs throughout our bodies, its effects are far-reaching. But even though cytosine modifications are ubiquitous, we’re still not quite sure what their biological function is.
Jack says, “Mounting evidence suggests hmC is involved in controlling how genes are expressed. It might fine tune the regulation of transcription, from DNA to RNA. But we need to understand a lot more about how hmC and other modifications are distributed in our genomes before we can fully decode their biological function.”
To zoom in on DNA’s inner workings, the team used mouse embryonic stem cells.
Stem cells are pluripotent – meaning they can develop into many kinds of cells.
Scientists first isolated these cells in the 1980s, when Cambridge’s Martin Evans and Matt Kaufman made a culture that suspended their developmental progression. Evans and Kaufman’s cells could multiply while keeping their pluripotency. Their culture became the reference point for labs around the world.
Researchers still need to keep a close eye on these cells: they’re so full of potential, they’re liable to start differentiating and move away from their pluripotent state. Sometimes researchers let the stem cells do this, for good reason – in this way, they can make ‘organoids’, like Madeline Lancaster’s ‘mini brains’.
Jack explains, “Early in embryonic development, cells divide and form a cluster, called a blastocyst. The stem cells we’re examining originate from inside the blastocyst. At that point, they can become pretty much any other cell.”
Jack thinks that cytosine modifications prepare embryonic stem cells for their manifold development, somehow telling them which body cells to become, and to remember their roles.
“HmC may prime cells to go down certain developmental pathways,” Jack says.
Researchers are searching widely for clues as to hmC’s function. HmC levels are at their highest in neurons – the cells that send and receive impulses throughout our nervous system. In forms of cancer, Alzheimer’s and Parkinson's, hmC functioning is impaired or dysregulated, suggesting a close involvement in these diseases. One speculative theory ties hmC to slowing down the rate of mutations, perhaps as a response to viruses lurking in our genetic code.
DNA can encode information in two ways: genetic information (the sequence of the bases A, C, G, T, light grey) and modifications, such as mC and hmC (yellow and red), which influence how genetic information is interpreted.
DNA can encode information in two ways: genetic information (the sequence of the bases A, C, G, T, light grey) and modifications, such as mC and hmC (yellow and red), which influence how genetic information is interpreted.
Unorthodox approaches
While Jack focuses on chemical annotations, current members of Shankar's group probe other unconventional aspects of DNA for information.
PhD student Zixuan Wang studies knots in the genome where the sequence is rich in guanine (the ‘G’ in ACGT this time, instead of Jack’s ‘C’).
At certain points, DNA bunches into a compact, four-stranded bundle rather than lying flat as the familiar double helix. These knots tend to form at the control panels of genes that are switched on, and they crop up in unusual abundance in cancer cells.
Zixuan's recent work targets the guanine knots, using them as an anchoring point to explore how DNA and proteins interact in their natural environment. She’s designed a molecule that can sit on the knots and flag particular proteins for destruction, including several that are known to drive cancer.
Both Jack and Zixuan hope their work leads to eventual targets for new therapies. Jack believes decoding hmC could help track how cancers develop and progress; Zixuan sees dysregulated DNA-protein interactions as a promising point of intervention.
We are just beginning to understand the full picture of how our body organises itself and why that organisation goes wrong. Breakthroughs like SCoTCH-seq let us interpret the hidden language of our genome, and the many combinations into which it can be remixed.
Published on 6 August 2026.
Words and images: Liam Morgan.
The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License