How Folding@home is mobilizing a million PCs for a coronavirus cure

As the novel coronavirus COVID-19 spreads around the globe, killing tens of thousands of people as of this writing and disrupting our daily lives in ways large and small, distributed computing project Folding@home is leading the search for a cure. Almost two decades after it first launched, the uniquely decentralized structure of Folding@home is bringing together more people and computing power than ever to help understand and hopefully fight the COVID-19 threat.

Folding@home is one of the longest-running distributed computing projects in the world, and arguably the most prominent. For nearly 20 years, the initiative, originally started out of Stanford University, has harnessed the power of computers all around the world to run simulations of protein folding in an effort to better understand maladies such as cancer, Alzheimer’s disease, Ebola—and now, coronavirus. Anyone can chip in, and every bit helps.

In late February, Folding@home—now based out of Washington University in St. Louis—announced that it was pivoting resources towards coronavirus research. On March 19, director Greg Bowman revealed on Reddit that the project had seen a 1,200% increase in volunteers, breaking the 400,000 device mark. By the end of March, that tally had surged past a million devices, setting a new peak for the project.

“As long as I don't think about why they're jumping in, it's exhilarating,” Bowman tells Decrypt. “We've gone from—3.5 weeks ago—having 30,000 volunteers in Folding@home, which was already great and let us do a lot of really valuable science… but now we have a million devices running Folding@home. It's estimated that we're the first to break the exaflop barrier, so the aggregate compute power we have is estimated to be more than the top 100 supercomputers combined.”

“It's a really cool tool to bring to bear on this very important problem,” he continues, “and it's giving us the opportunity to try as many things as we can think of in parallel to see if anything bears fruit sooner rather than later. Of course, we can't guarantee what results we'll get and what the time scales will be, but we've certainly had related success with other systems—so we're hopeful."

How is Folding@home helping coronavirus research?

Bowman’s group was already focused on other infectious diseases when a member of his team began simulating one of COVID-19’s proteins. Quickly it dawned on them that they should be exploring all of the disease’s proteins to try to understand where potential remedies could be most effective. Bowman explains that one part of the process is simulating how certain chemical compounds bind most strongly to proteins in particular sites, helping their partners narrow down the options while developing remedies.

“This is really valuable, because our experimental collaborators—they can do a lot, but they only have finite resources, so they can't buy every possible chemical that you could potentially make and synthesize those and do the experiments,” says Bowman. “What we're able to do is help them prioritize compounds that are more likely to work than others."

Folding@home began in late 2000 and has continued on over the years, just moving under Bowman’s direction in 2019. He says that the original idea for the project was to figure out how the process of protein folding leads to the various maladies they were investigating.

"It's as if you took all the components of your car and laid them out in a line on the street, and they spontaneously on their own, without anyone assembling them, sprung into place and formed a functioning car,” he explains. “It's like: Wow, how does that happen? And so the idea was that if we could solve problems like that, then we'd be in a position to tackle all kinds of other more immediate, biomedically relevant problems."

Trying to find out where drugs could potentially bind to a protein in a disease is something that Bowman’s group was heavily focused on with Ebola before pausing due to coronavirus. Now it’s something they’d like to apply to the current pandemic.

“Oftentimes, proteins don't have any attractive sites for a drug to bind tightly. But when we go in with our simulations and look at what the proteins do, besides the most usual structure that you can see experimentally, we often see the spontaneous formation of appealing-looking sites for drugs to bind that we call cryptic pockets,” says Bowman. “We're set up then to experiment with tests to see if these pockets are real, and test whether targeting them has the desired functional impact on the proteins—and then to start designing drugs to bind with them.”

Bitcoin miners join the coronavirus fight

Some pockets of the crypto mining community have been a key part of Folding@home’s recent surge in computing power. Both blockchain and Folding@home are built on the notion of decentralization, and the call to action to help Folding@home complete more and more simulations has convinced some avid miners to pivot their resources towards a coronavirus cure.

"It's been really cool. I've been really happy with that community. A lot of them have shut down mining operations and redirected computing power to us,” says Bowman. “We have groups that effectively run cloud services that backfill with cryptocurrency mining, so they've been backfilling with Folding@home instead. It really affects their finances, so it's quite a generous move.”

Bowman pointed to CoreWeave, the Ethereum mining company that redirected more than 6,000 GPUs towards Folding@home, as co-founder and CTO Brian Venturo told CoinDesk. At the time, the company was reportedly contributing more than half of the total GPU power focused on COVID-19 simulations.

Numerous Tezos mining groups have also pointed their resources towards Folding@home in recent weeks for a computing competition, plus the Bitcoin subreddit organized a team for folding. “I feel it would make Satoshi proud to know that the little community he seeded grew into something that would be brave enough to take on the monsters of this world, in more ways than one,” wrote Reddit user Fiach_Dubh.

Meanwhile, the long-running Curecoin initiative—in which participants earn tokens for folding—was designed for this exact kind of challenge. These are just a few examples, but many more miners seem to be using their powerful rigs to contribute to a cure.

Ultimately, whether it’s an individual crypto miner or a group, a PC gamer with a souped-up rig, a university tapping its currently-idle computers, or anyone else with a compatible device, it all helps Folding@home run more simulations and explore more possible breakthroughs.

Everyone can play a part in fighting the coronavirus and flattening the curve by staying at home, honoring social distancing regulations, and practicing good hygiene. But for those who feel aimless amidst this pandemic and don’t know how to contribute, Folding@home offers participants an opportunity to be a part of something meaningful. Again, every bit counts.

"It's really cool to see people coming together and trying to do what they can to help out,” says Bowman. “I think that's one of the cool things that we can offer, is the potential to do something proactive to fight against this virus. It's a little more gratifying than hand washing and social distancing.”

The future of Folding@home after COVID-19

The coronavirus will continue to be Folding@home’s focus for the foreseeable future, and the process of developing a legitimate vaccine or treatment could be extensive. With so much additional computing power right now, the team hopes to ask and try to answer a lot more different types of questions, and explore the way that the viral proteins “hijack human cellular machinery,” adds Bowman.

Ultimately, he stresses that while the dramatic increase in volunteers has been amazing, it’s the bedrock of some 30,000 active pre-coronavirus volunteers that kept the project going through other pursuits and allowed them to rapidly shift gears to focus on COVID-19. Bowman hopes that the volunteer pool stays sizable to help understand and fight other diseases and ailments in the future once this current need subsides.

“One of my big hopes is that people will recognize that we were only able to jump on this so quickly and start making some impact because we already had these 30,000 people that were participating in Folding@home before, and helping with our research on cancer and Alzheimer's disease and Ebola,” he says. “I hope people will appreciate that this is a general platform for tackling important human maladies, and that lots of them will stick around and keep helping us after the immediacy of the COVID-19 pandemic has hopefully been alleviated."