By combining genomics, data science and artificial intelligence, researchers are transforming how pandemics are detected, tracked and managed, helping governments respond faster and more effectively to emerging infectious diseases.
When the Covid-19 pandemic struck, infectious disease specialists from the Oxford Martin School were ready to arm governments with crucial genetic information showing how the virus was evolving and how fast it might spread.
The Oxford Martin Programme on Pandemic Genomics had begun in 2018, integrating mathematical and statistical disease modelling with advances in sequencing large volumes of genetic data. Previous outbreaks of swine flu, Zika and Ebola had allowed the team to demonstrate how important large-scale whole genome sequencing was, and how it could help governments make better-informed decisions about interventions such as travel restrictions and vaccine rollout.
“Standard mathematical epidemiology doesn’t tell you who infected whom, what mutations are appearing and how significant those mutations might be,” says Professor Oliver Pybus, one of the programme’s Directors.
“Standard mathematical epidemiology doesn’t tell you who infected whom, what mutations are appearing and how significant those mutations might be”
The scale of the pandemic meant scientists were soon faced with analysing millions, rather than thousands, of genomes, and the team had to create new analysis techniques from scratch. Emergency funding from the UK government and the Wellcome Trust enabled labs across the UK to work together, sharing data and generating rapid insights into how the disease was behaving
From analysing the first wave of the outbreak, the Oxford Martin School team found the virus was mainly entering the UK not from China, as widely assumed, but from Europe, with thousands of independent transmission events driven in part by the February 2020 half-term holiday.
The Oxford Martin School team were also co-discoverers of the first ‘variant of concern’, Alpha. “Alpha was an evolutionary leap, rather than an evolutionary step,” says Professor Pybus. “It had a dozen new mutations, rather than just one or two, and they helped make the variant far more transmissible.”
By integrating travel data with genomics, the team showed how large-scale population movements drove the spread of the virus. They also co-established the technical naming system for Covid-19 variants, providing a shared scientific language that supported global collaboration during the pandemic.
Advancing epidemic intelligence
Building on this work, the programme is now developing new approaches that combine genomic data with epidemiological and mobility data to better understand how diseases spread in real time. “Increasingly we are using data-driven approaches to disease surveillance and modelling, tracking disease spread across populations, inform and evaluate intervention strategies” says Professor Moritz Kraemer, one of the programme directors.
Recent research shows that combining pathogen genomes with case data provides a more accurate picture of how infections grow and decline during an outbreak.
The team is also at the forefront of applying artificial intelligence to infectious disease research. A major review published in Nature highlights how AI and machine learning can transform epidemic forecasting, surveillance and decision-making, particularly when combined with large-scale genomic datasets.
This enables:
- Faster detection of emerging threats
- More accurate tracking of disease spread
- Better-informed decisions on when and how to intervene during an outbreak
Smarter surveillance and better data
The programme is also exploring how genomic surveillance can be made more targeted and effective during an outbreak. Using “active learning”, researchers can identify where sequencing efforts are most needed, helping to detect patterns of spread more efficiently and make better use of limited resources.
At the same time, the team is also examining the limits of genomic data, showing how gaps in sampling can distort our understanding of how viruses spread between regions, particularly when distinguishing between locally transmitted and imported cases.
Beyond Covid-19
Beyond Covid-19, the team has applied its methods to other global health challenges. Research published in Science showed how pandemic-related travel restrictions reshaped the global circulation of seasonal influenza, reducing international spread and disrupting long-established patterns of virus movement.
Informing policy and global preparedness
The programme’s research is also shaping policy and practice at national and international levels. Professor Oliver Pybus serves on the UK government’s Scientific Pandemic Influenza Modelling group (SPI-M) and on a UK Health Security Agency advisory group on avian influenza. He also contributed to Operation Pegasus, a national pandemic response simulation exercise.
“Thanks to the work of our team and of many other colleagues worldwide there’s now a much better understanding within governments of what virus genomics can contribute.”
Internationally, Professor Moritz Kraemer contributes to World Health Organization (WHO) expert groups developing frameworks for genomic data sharing and pandemic preparedness. This work has informed global guidance on how pathogen genomic data should be shared and used to support surveillance and public health action, helping to strengthen coordinated responses to epidemic and pandemic threats.
Preparing for future pandemics
Together, this work is helping to build a new, data-driven framework for pandemic preparedness. It combines genomics, artificial intelligence and real-time data to support faster, more informed responses to future outbreaks.
“Thanks to the work of our team and of many other colleagues worldwide there’s now a much better understanding within governments of what virus genomics can contribute.” — Professor Oliver Pybus, Professor of Evolution & Infectious Disease
