Kei Sato was looking for his next big challenge five years ago when it smacked him — and the world — in the face. The virologist had recently started an independent group at the University of Tokyo and was trying to carve out a niche in the crowded field of HIV research. “I thought, ‘What can I do for the next 20 or 30 years?’”
He found an answer in SARS-CoV-2, the virus responsible for the COVID-19 pandemic, that was rapidly spreading around the world. In March 2020, as rumours swirled that Tokyo might face a lockdown that would stop research activities, Sato and five students decamped to a former adviser’s laboratory in Kyoto. There, they began studying a viral protein that SARS-CoV-2 uses to quell the body’s earliest immune responses. Sato soon established a consortium of researchers that would go on to publish at least 50 studies on the virus.
In just five years, SARS-CoV-2 became one of the most closely examined viruses on the planet. Researchers have published about 150,000 research articles about it, according to the citation database Scopus. That’s roughly three times the number of papers published on HIV in the same period. Scientists have also generated more than 17 million SARS-CoV-2 genome sequences so far, more than for any other organism. This has given an unparalleled view into the ways in which the virus changed as infections spread. “There was an opportunity to see a pandemic in real time in much higher resolution than has ever been achievable before,” says Tom Peacock, a virologist at the Pirbright Institute, near Woking, UK.
On supporting science journalism
If you’re enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Now, with the emergency phase of the pandemic in the rear-view mirror, virologists are taking stock of what can be learnt about a virus in such a short amount of time, including its evolution and its interactions with human hosts. Here are four lessons from the pandemic that some say could empower the global response to future pandemics — but only if scientific and public-health institutions are in place to use them.
Viral sequences tell stories
On 11 January 2020, Edward Holmes, a virologist at the University of Sydney, Australia, shared what most scientists consider to be the first SARS-CoV-2 genome sequence to a virology discussion board; he had received the data from virologist Zhang Yongzhen in China.
By the year’s end, scientists had submitted more than 300,000 sequences to a repository known as the Global Initiative on Sharing All Influenza Data (GISAID). The rate of data collection only got faster from there as troubling variants of the virus took hold. Some countries ploughed enormous resources into sequencing SARS-CoV-2: between them, the United Kingdom and the United States contributed more than 8.5 million (see ‘Viral genome rally’). Meanwhile, scientists in other countries, including South Africa, India and Brazil, showed that efficient surveillance can spot worrying variants in lower-resource settings.
In earlier epidemics, such as the 2013–16 West African Ebola outbreak, sequencing data came in too slowly to track how the virus was changing as infections spread. But it quickly became clear that SARS-CoV-2 sequences would arrive at an unprecedented volume and pace, says Emma Hodcroft, a genomic epidemiologist at the Swiss Tropical and Public Health Institute in Basel. She works on an effort called Nextstrain, which uses genome data to track viruses, such as influenza, to better understand their spread. “We had developed so many of these methods that, in theory, could have been very useful,” Hodcroft says. “And all of a sudden, in 2020, we had an opportunity to put up and show up.”
Initially, SARS-CoV-2 sequencing data were used to trace the spread of the virus at its epicentre in Wuhan, China, and then globally. This answered key early questions — such as whether the virus spread largely between people or from the same animal sources to humans. The data revealed the geographical routes through which the virus travelled, and showed them much more quickly than could conventional epidemiological investigations. Later, faster-transmitting variants of the virus started appearing, and sent sequencing labs into hyperdrive. A global collective of scientists and amateur variant trackers trawled through the sequence data constantly in search of worrying viral changes.
“It became possible to track the evolution of this virus in tremendous detail to see exactly what was changing,” says Jesse Bloom, a viral evolutionary biologist at the Fred Hutchinson Cancer Center in Seattle, Washington. With millions of SARS-CoV-2 genomes in hand, researchers can now go back and study them to understand the constraints on the virus’s evolution. “That’s something we’ve never been able to do before,” says Hodcroft.
Viruses change more than expected
Because no one had ever studied SARS-CoV-2 before, scientists came with their own assumptions about how it would adapt. Many were guided by experiences with another RNA virus that causes respiratory infections: influenza. “We just didn’t have much information about other respiratory viruses that could cause pandemics,” says Hodcroft.
Influenza spreads mainly through the acquisition of mutations that allow it to evade people’s immunity. Because no one had ever been infected with SARS-CoV-2 before 2019, many scientists didn’t expect to see much viral change until after there was substantial pressure placed on it by people’s immune systems, either through infections or better yet, vaccination.
The emergence of faster-transmitting, deadlier variants of SARS-CoV-2, such as Alpha and Delta, obliterated some early assumptions. Even by early 2020, SARS-CoV-2 had picked up a single amino-acid change that substantially boosted its spread. Many others would follow.
“What I got wrong and didn’t anticipate was quite how much it would change phenotypically,” says Holmes. “You saw this amazing acceleration in transmissibility and virulence.” This suggested that SARS-CoV-2 wasn’t especially well adapted to spreading between people when it emerged in Wuhan, a city of millions. It could very well have fizzled out in a less densely populated setting, he adds.
A healthcare worker uses a pipette to process Covid-19 test samples at the SpiceHealth Genome Sequencing Laboratory set up at the Indira Gandhi International Airport in New Delhi, India, on January 14, 2021.
T. Narayan/Bloomberg via Getty Images
Holmes wonders, also, whether the breakneck pace of observed change was merely a product of how closely SARS-CoV-2 was tracked. Would researchers see the same rate if they watched the emergence of an influenza strain that was new to the population, at the same resolution? That remains to be determined.
The initial giant leaps that SARS-CoV-2 took came with one saving grace: they didn’t drastically affect the protective immunity delivered by vaccines and previous infections. But that changed with the emergence of the Omicron variant in late 2021, which was laden with changes to its ‘spike’ protein that helped it to dodge antibody responses (the spike protein allows the virus to enter host cells). Scientists such as Bloom have been taken aback at how rapidly these changes appeared in successive post-Omicron variants.
And that wasn’t even the most surprising aspect of Omicron, says Ravindra Gupta, a virologist at the University of Cambridge, UK. Shortly after the variant emerged, his team and others noticed that, unlike previous SARS-CoV-2 variants such as Delta that favoured the lower-airway cells of the lung, Omicron preferred to infect the upper airways. “To document that a virus shifted its biological behaviour during the course of a pandemic was unprecedented,” says Gupta.
Omicron’s preference for upper airways probably contributed to its clinical mildness — its relatively low virulence — compared with previous iterations. But that shift is hard to disentangle from the fact that Omicron struck after much of the world had begun to establish some immunity, says Bloom, and there is evidence that Omicron was just as nasty as the version of SARS-CoV-2 that emerged in Wuhan.
And although Omicron and its offshoots were milder than Alpha, Beta and Delta, those had all proved more virulent than the lineage they replaced, toppling the idea that the virus would evolve to be less deadly. “The idea that there’s some law of nature that says that a virus is going to rapidly lose its virulence when it jumps into a new host is incorrect,” Bloom says. It’s an idea that never had much buy-in with virologists anyway.
One of Sato’s big fears is that a drastically different SARS-CoV-2 variant will emerge and overcome the immunity that stops most people becoming severely ill. He worries that the result could be disastrous.
Chronic cases could reveal insights
Before Gupta turned his attention to SARS-CoV-2, his focus was HIV, which is ordinarily a lifelong infection. As a clinician, he had treated the second person ever cured of HIV through a blood stem-cell transplant. But his research group studied how antiretroviral drug resistance evolves over the course of months and years in people.
Most scientists presumed that, unlike HIV or other long-term infections, respiratory viruses such as SARS-CoV-2 were acute, and those who survived their infections cleared the virus in a matter of days. Longer-term infections occur in influenza, but they seem to be an evolutionary dead end. The virus adapts to survive in the host, not to spread to others.
But in late 2020, Gupta characterized a 102-day SARS-CoV-2 infection in a man in his 70s with a compromised immune system. The infection was ultimately fatal. In the man’s body, the virus developed a high number of spike-protein changes. Many of these would also be observed in worrying variants, including the Alpha variant that sent case counts rocketing and prompted another wave of lockdowns in late 2020 and early 2021.
The man’s case didn’t give rise to any widespread variant, but it gave Gupta, with his HIV evolution background, the idea that chronic infections could be a source of the drastic evolutionary leaps that characterized SARS-CoV-2 variants of concern. “We didn’t have the preconceptions the flu field had of what respiratory viruses do,” he says.
Alex Sigal, a virologist at the Africa Health Research Institute in Durban, South Africa, had a similar idea when another variant, called Beta, was identified in his country. South Africa has a high rate of HIV infections — many of which go untreated — and Sigal wondered whether it was more than a coincidence that Beta seemed to have emerged where there were high numbers of people who were immunocompromised.
Omicron — which was first detected by scientists in Botswana and South Africa — strengthened the case that long-term infections are a source of variants. Omicron was also littered with spike mutations that had been observed in people who were immunocompromised. Researchers have observed similar evolution by tracking ‘cryptic’ SARS-CoV-2 lineages identified in wastewater sampling but not seen elsewhere.
No one has yet identified the precise source of Omicron or any of the major variants, but most scientists now think that they begin in people with chronic infections, during which the virus has time to string together otherwise improbable combinations of mutations that evade immunity and boost transmission (exactly how is an active area of research). Scientists, including Sigal, have begun studying immunocompromised individuals, including people with untreated HIV infections, to better understand the characteristics that can give rise to the viral evolution observed in variants such as Omicron.
Researchers are also now asking whether chronic infections are important to the evolution of other pathogens, including the viruses that cause mpox, chikungunya, Ebola and RSV, a common respiratory virus that can cause severe disease in young children and older people. “This is something that is a paradigm-shifting observation from COVID-19, and we’ll now be looking for this in future pandemic viruses,” says Gupta.
A responsive way of doing science
Sato uses the term ‘responsive science’ to describe how his lab operated during the pandemic. As soon as a worrying new variant was observed, researchers around the world and a number of highly skilled non-scientists started scouring the data. Sato’s team worked around the clock characterizing variants — learning about their capacity to dodge immunity or spread from cell to cell — and churning out data in days or weeks, rather than years. When another variant emerged, the cycle repeated.
“This was one of the first times where evolutionary biology became an applied science,” says Bloom. His lab conducted ‘deep mutational scanning’ experiments that probed the effects of tens of thousands of potential, predicted viral changes.
The rush to study SARS-CoV-2 delivered effective vaccines, therapeutics, such as monoclonal antibodies, and actionable insights into the virus’s spread. “People’s mindsets changed,” says Sigal. If the same levels of data sharing, collaboration and urgent investment became common in other spheres such as cancer biology, he argues, it could save more lives.
Susan Weiss, a virologist at the University of Pennsylvania’s Perelman School of Medicine in Philadelphia who has studied coronaviruses since the late 1970s, says that the successful race to develop vaccines, especially those based on messenger RNA, was probably the most important lesson from the pandemic. But beyond that, she questions whether the rush to study SARS-CoV-2 created a knowledge base that scientists studying the basic biology of other coronaviruses can build on. Many labs have moved on from SARS-CoV-2. “I don’t know a lot of people who stuck with it,” Weiss adds.
Sato’s lab is still focused on SARS-CoV-2. Part of the move away from the virus is due to the lack of urgency — and long-term funding. SARS-CoV-2 sequencing has levelled off: last year, fewer than 700,000 sequences were added to the GISAID repository.
The experience of studying SARS-CoV-2 so intensely also left many scientists burned out, says Peacock. “It’s quite soul-destroying, because you just end up feeling like a production line rather than a science unit doing hypothesis-driven science.” He’s now working on another potential pandemic-causing virus: H5N1 avian influenza.
Many researchers are now asking what is the right level of sequencing for SARS-CoV-2 — and other human and animal pathogens — given scant resources and unknown threats. Peacock hopes for a deep reserve of capacity. “Can we use that existing infrastructure to have a peacetime way of running things, but then can quickly ramp up to a wartime one?” asks Peacock.
Hodcroft would like to see more sequencing to monitor changes in viruses that people regularly encounter, such as RSV, seasonal coronaviruses or human metapneumovirus, which tend to cause mild respiratory infections. Paying close attention to diverse pathogens will broaden people’s understanding of where future threats might lurk. The virus behind the next pandemic could hold even bigger surprises than SARS-CoV-2 did.
Yet some researchers worry that the opportunities presented by SARS-CoV-2 research are now being squandered, particularly in the United States after the election of President Donald Trump. With cuts to federal funding for public health and research, the intention to pull out from the World Health Organization and other upheavals, his administration has limited scientists’ ability to track and respond to infectious disease and to share information, they say. “If you look at the policies that are being implemented, we’ve actually gone backwards,” says Angela Rasmussen, a virologist at the University of Saskatchewan in Saskatoon, Canada.
In the early days of the pandemic, it seemed as if politicians were open to the lessons to be learnt from SARS-CoV-2. In 2020, world leaders, including those in the United States, looked ready to establish a global pathogen surveillance network, Holmes says. “The politics have mired it down,” he says. “We’re actually in a worse place in terms of pandemic prevention than we were before the pandemic started.”
This article is reproduced with permission and was first published on March 12, 2025.
