Tell us about the idea behind this theme issue and how it came about.
This theme issue stemmed from the need to highlight the scientific contributions that the Royal Society RAMP initiative has made over the last two pandemic years. To do this, we invited a diverse range of academics, most of whom contributed to the RAMP initiative, to showcase their work on modelling the transmission of COVID-19, and to discuss the technical issues identified during the current epidemic that would be relevant when modelling future ones.
The idea behind the theme issue was not just to use one set of tools (for example, SEIR modelling plus Approximate Bayesian Computation or Monte Carlo Micro Simulation calibration techniques, which have been the go-to modelling method for infectious disease spread), but to instead dive into different disciplines and methods that offer an alternative way of modelling the pandemic.
What was the Royal Society RAMP initiative?
The coronavirus pandemic highlighted the importance of scientific modelling becoming a tool for informed policy advice on the control of the spreading of the SARS-CoV-2 variants. In February 2020, the UK already had a number of strong, established epidemiology groups and an independent national advisory body, SPI-M, which had a lot of expertise in modelling infectious disease spread. But the scale of the growing pandemic was a larger task than the existing epidemiology groups could handle.
The RAMP initiative offered an additional set of diverse modelling expertise to support the pandemic modelling community already working on COVID-19. RAMP was a novel and experimental way of organising science during an emergency. Designed and led by scientists, it operated on a volunteer basis and without a budget supplementing the SPI-M work. The most pressing concern was to augment the existing groups, adapting and accelerating existing code to the specific needs and data streams associated with the pandemic. A stream within RAMP was also reviewing the ever-growing number of preprints that started to emerge, which were invaluable during the pandemic. Overall, RAMP provided more than 15 years of postdoctoral level support to the existing UK epidemiology community.
What do you think is the most exciting idea discussed in the papers?
As someone who has been actively modelling the COVID-19 pandemic, it was exciting to see the breadth of new modelling techniques - often borrowed from different disciplines - that have been developed at pace over the last two years. For example, even though agent-based models (ABMs) have always existed as an alternative to SEIR population-based modelling techniques for infectious disease spread, the need to model contact-tracing and the behaviour of individuals rather than population groups, really highlighted their importance in pandemic outbreak control. This is why this issue has four featured papers that utilise ABMs – Sanz-Leon et al., Hinch et al., Li et al. and also one led by myself - each showcasing different methods and applications to different dates and settings.
One of the gems of this issue is also the fact that we had academics from many different scientific backgrounds (e.g., theoretical physics, mathematics, biology, statistics) all contributing to it.
Did you learn anything new when editing the papers?
I appreciated for the first time how methods using emulators (which might be described as "machine learning") can be applied to data-rich applications. Interestingly, the application by Vernon et al. was not to understand the epidemic data itself but to model the inputs and outputs of the related agent-based model. I suspect that this approach will become more important in the future as we become less reliant on individual model assumptions and outcomes, but strive to better understand what their inputs and outputs tell us in general.
How was your experience of being a Guest Editor on Phil Trans A?
I very much enjoyed the experience as a Guest Editor at Philosophical Transactions A. Being a mathematician, it was a great honour and a personal accomplishment to have served as a Guest Editor for this historical journal, where Sir Isaac Newton published his landmark paper on the nature of light and colour in 1672. Furthermore, this journal has also extended the natural sciences into public health with the publication of Hans Sloane’s account of inoculation with smallpox in 1755. Interlinking technical work to improve public health is very much at the heart of our theme issue and I feel that we are continuing the legacy of the journal with it!
Furthermore, I have thoroughly enjoyed working closely with my very involved co-editors: William Waites from Strathclyde University, and Graeme Ackland from University of Edinburgh. Their theoretical physics backgrounds and my mathematics background have complemented each other in compiling an inter-disciplinary issue which we hope will remain relevant in years to come. I am also very grateful for the support and the excellent interaction with Alice Power who, as the Commissioning Editor of the journal, made our editing task enjoyable and was instrumental in the timely and appropriate completion of this issue. It is also worth noting that we have been editing this special issue during the COVID-19 pandemic, which on its own has been challenging and would have been much delayed had it not been for the timely responses of the invited authors – many of whom were simultaneously modelling the pandemic and advising policy decision makers. Finally, I am very grateful for the timely and efficient responses from the voluntary anonymous reviewers without whom this theme issue wouldn’t have been possible.
What is the future for research in this area?
Mathematical and scientific modelling has played a pivotal role in producing timely and responsive scientific advice for policy decision makers. But over the course of the pandemic, different issues have been identified that need to be incorporated within modelling, such as the use of genomic data, better methods for analysis and utilisation of complex data, and improved methods for robust calibration of complex models. The RAMP initiative paved the way for accessing methodologies that are routinely used in other disciplines to be utilised for epidemiological nowcasting and forecasting. It is important to expand this cross-disciplinary approach in the future.
I strongly feel that the template of the RAMP initiative shows an important way of linking ready and willing volunteers to work on pressing issues and it is a great example of how ‘crowd-thinking’ and collaborative working can be timely, fruitful and very impactful. This model for science is one which can be deployed again in many future emergencies - not just pandemics. There are a number of scientists and academics who just love doing science and the thought of making an immediate, tangible, positive impact on society is enough to inspire us to work for free.
Tell us a bit about your own research.
In my research I utilise mathematics and statistics together with computational techniques to develop mathematical models combined with complex real-life datasets to answer questions in public health. Pre-COVID-19 I had been developing and using mathematical models for transmission and control of infectious diseases such as HIV and influenza. Since early 2020 I have been actively modelling the COVID-19 pandemic across different settings, including the UK, and advising scientific and policy decision makers. One focus of my COVID-19 research has been related to schools and SARS-CoV-2 transmission – with a number of my recent papers assessing optimal conditions to reopen schools after the different COVID-19-induced lockdowns.
Read the theme issue '