Unravelling the secrets of DNA sequences at the molecular level

Researchers in the Department of Physics have developed a new technique that can observe thousands of individual DNA molecules at work and read the exact sequence of each one. This innovation could help scientists better understand how DNA sequence controls molecular functions and biological processes. The method was developed by Dr Jagadish Prasad Hazra, Dr Rebecca Andrews and Professor Achillefs Kapanidis with colleagues at Oxford's Kavli Institute for Nanoscience Discovery and the Department of Chemistry. Their work has been published in Nature Communications.

DNA carries the instructions for life, but even small differences in its sequence can change how it interacts with other molecules. Until now, it has been difficult to watch what happens to individual DNA molecules and then work out the exact sequence of each one, making it hard to understand exactly how sequence controls function.

The new method, called Single-molecule Phenotyping and In-situ Sequencing (SPIN-Seq), allows researchers to observe single DNA molecules as they interact with proteins or other molecules, and then read the sequence of that same molecule on the same surface and instrument, directly linking what each molecule does (its 'phenotype') to the DNA letters it contains (its 'genotype').

The core novelty of the method is the ability to sequence a single DNA molecule in situ, immediately after a functional assay has been carried out on it. It is also the first single-molecule method that uses only DNA, rather than protein, to sequence DNA, reading up to 5 bases with high accuracy across libraries of up to 1,024 DNA sequences.

Using SPIN-Seq, the team dissected how a transcription factor interacts with its target sequence, and revealed how DNA sequence controls the pausing and reaction pathways of ribonucleic acid (RNA) polymerase (the enzyme responsible for reading genes and copying their information into RNA) during the early stages of transcription.

'Single-molecule studies are very powerful, but typically address only one DNA sequence at a time; that's why it is very hard to understand how the bases in the DNA control how proteins read the different "codes" hidden in DNA molecules,’ explained Professor Kapanidis. ‘Back in 2016, we started dreaming of single-molecule ways where we can work with DNA libraries to tackle these hidden mechanisms in a "holistic" way, and learn the "rules" of the game. After years of hard work, we are now very excited to see our dreams becoming a reality, and offering many new ways to learn and exploit the link between sequence and function.'

The new technology is part of a new family of powerful multiplexed single-molecule methods that connect sequence to function for libraries of molecules, covered in a recent Science review. The technology behind SPIN-Seq has been supported by Oxford University Innovation (OUI), the University's research commercialisation company, which has helped the team develop its patent protection.

Beyond helping scientists understand how DNA sequence shapes molecular interactions and biological processes, SPIN-Seq could have practical applications too. By rapidly identifying molecules with useful behaviours from large libraries, the technique could provide new ways to search for potential drugs, diagnostic sensors and other useful molecules. In the fast-moving landscape of AI- and machine-learning-enabled biology, SPIN-Seq opens a powerful route toward dynamics- and kinetics-guided discovery, where molecular behaviour can be learned, modelled, and ultimately used to engineer the next generation of DNA–protein interaction systems. The large datasets it generates could also provide valuable training data for AI models, helping them learn how molecular structure relates to function, and ultimately to predict and design new molecules with desired properties.

Unraveling single-molecule reactions via multiplexed in-situ DNA sequencing, JP Hazra, et al., Nature Communications, 29 August 2026.