The revival of the "lab leak" theory of COVID-19 — that it started with a virus that originated in a laboratory — is drawing new attention to the dangers of scientific experiments that enhance pathogens to study them.

Why it matters: So-called "gain-of-function" research can be invaluable in predicting future pandemic threats, but it can also generate risks by introducing new, potentially more dangerous viruses.

Driving the news: Last week, a group of leading researchers called for a renewed investigation into the origins of COVID-19 in a letter published in the journal Science.

• It was the most prominent sign that the initial explanation for the emergence of SARS-CoV-2 — that it spilled over from an animal source — is coming into doubt.
• There's no bulletproof evidence for either hypothesis yet, and the basic fact that past outbreaks have repeatedly been started by animal spillovers puts a higher burden of proof on the possibility of a lab leak.
• But as Harvard epidemiologist Marc Lipsitch — a signatory to the Science letter — told WBUR last weekend, "If this did come from a laboratory, then that really has implications for the kind of work that we should be doing and the kind of safeguards that need to be in place if we're going to do that work."

How it works: Gain-of-function research involves purposefully enhancing pathogens in a lab to better understand them.

• Proponents of such research argue it is one of the best ways to prepare for future pandemics, allowing scientists to identify potential genetic changes in pathogens that could make them more transmissible or virulent, and to design and stockpile vaccines and treatments.
• Opponents say the research itself poses an unacceptable risk by bringing into the world more dangerous viruses that wouldn't have otherwise existed — potentially triggering pandemics if they escape in a lab accident.

The backstory: Concerns around gain-of-function research, prompted by work with the H5N1 avian flu virus, led the National Institutes of Health to put a funding pause in 2014 on such studies involving dangerous viruses like flu or SARS. The pause was lifted three years later after the creation of an oversight framework.

Like all low-probability/high-consequence events, hard data on both the true risks and benefits of gain-of-function research is hard to come by, which makes it difficult to estimate the scale of the threat.

• In 2014, Lipsitch and Tom Inglesby of the Johns Hopkins Center for Health Security estimated the risk of a pandemic caused by an accidental release from a high-security biolab doing such work with a transmissible flu virus over a single year at between 1 in 10,000 and 1 in 100,000.
• At the lower level, that's less than the expected risk of a random pedestrian dying in a traffic accident in a given year. Should a pandemic spill out of a lab, however, the result wouldn't be one person dying, but potentially millions.

Yes, but: An accidental human-made pandemic requires not just the creation of a more dangerous virus in a lab, but a safety breakdown that allows it to escape — and some experts argue that gain-of-function skeptics have overestimated the likelihood of such safety lapses.

• Outlawing or even more tightly regulating gain-of-function work would carry unknown risks itself, as we might lose out on research that could protect us against future pandemics.

What's next: Barring a sudden shift toward transparency by the Chinese government, we may never know conclusively the true origins of COVID-19. But the risk from human-made pathogens is almost surely growing.

• More labs around the world are pursuing potentially risky virus research, and some of it is likely being done in less-than-optimal safety conditions — as may have been the case at the Wuhan Institute of Virology.
• New tools that make it cheaper and easier to write gene sequences raise the risk as well, prompting California to move on a bill this week that would require DNA synthesis companies to adopt what are currently voluntary protocols to screen for dangerous sequences.
• Whatever the ultimate results of the COVID-19 investigation, researchers should double down on lab safety, which has too often been an afterthought.

The bottom line: Should we discover that COVID-19 originated in research — however well-intentioned — a recent boost in public support for science could evaporate, which might be the greatest risk of all.