Opportunities and limitations for big data in addressing present and future pandemics
The use of data has played a central role in the COVID-19 pandemic, but it should come as no surprise that researchers are already looking at ways to improve data acquisition, management, and access.
EMBL's most recent Science and Society seminar, 'Harnessing Big Data to Monitor and Tackle Pandemics', featured a panel of three speakers who shared their experiences and conclusions about these topics.
Ilaria Capua, Director of the One Health Center of Excellence at the University of Florida, explained why the concept of 'circular health' could transform what we know about pandemics and how we address them. Enrico Bucci, Director of the System Biology programmes at Temple University in Philadelphia, discussed the limitations of big data and the idea of long-term pandemic forecasting.
And Rolf Apweiler, Director of EMBL's European Bioinformatics Institute (EMBL-EBI) and Senior Scientist, focused on the ways in which big data is most suited to answering specific questions about the pandemic. Ewan Birney, EMBL Deputy Director General and Director of EMBL-EBI, moderated the discussion and took questions from the audience.
Here are seven key messages to take away.
1. The COVID-19 pandemic is the most measured event in human history.
As the panellists acknowledge, this pandemic has generated and continues to generate a deluge of data on virus genomics, patient symptoms, COVID-19 test results, treatment approaches, safety and effectiveness of vaccines, and data related to geography, the environment, and animal vectors. Rather than presenting one clear picture, these are more like puzzle pieces on the floor, as Rolf describes them. The result has been disparate interpretations from different fields of science. The good news is that scientists have a lot of information - even about new viral variants.
2. Having a lot of data is not a panacea.
Even though researchers have tremendous amounts of information, there's a limit to what it can do. Described by Enrico as "deterministic chaos", the COVID-19 pandemic is a situation where we may know all the previous circumstances we've been in, but we still can't say what will happen next because the system is chaotic. In these situations, there's a limit to how accurately we can predict when the next wave will start or how long it will last. We also need to continue improving our ability to mine the data and to do so quickly.
3. Be critical of your data.
Researchers who work with only part of the data may only see part of the story. It's incumbent on researchers to be sceptical of their data, tracking back to evaluate its coherence and integrity.
4. Collaboration is crucial.
Scientific collaboration provides many benefits: most importantly, a broader perspective that can bring together scientific disciplines and provide a more meaningful understanding of risks of infection.
5. There's value in near-casting.
While there may be limitations to long-term forecasting, there's potential for near-casting (e.g. a 10-day forecast of the pandemic in your neighbourhood). After ten days, models can diverge a lot. And 'now-casting' is equally important, if not more important, says Ewan. Unfortunately, our understanding of 'now' is rather imprecise.
6. One of the challenges is that human beings don't see themselves as animals.
People live in the context of their environments, not in isolation. Ilaria outlined the idea of 'circular health', which aims to "co-advance what we know about humans, plants, and animals in one system". Collaborative, international research is central to gaining a better understanding of diseases that move between animals and humans, and the way this transmission is connected to the changing environments we share.
7. Policymakers and politicians need to know science.
The question-and-answer session highlighted the need for policymakers to understand science if they're to develop effective policies and communicate meaningful messages to the public. That can mean electing scientists and science-literate candidates.
Source: EMBL Heidelberg