You Can Help Fight Coronavirus by Giving Scientists Access to Your Computer
Stanford’s Folding@home is using distributed computing to help develop COVID-19 drugs
With over 100,000 confirmed cases, the race to find a vaccine and cure for COVID-19, the virus that is making its way around the world, is in full swing. But scientists from the Folding@home Consortium believe they can speed up this process with your help.
What is the Folding@home project?
Folding@home (FAH) is a project run by Stanford University’s Pande Lab that uses the idle resources of personal computers around the world to perform disease research. On February 27, the team behind the project announced they are taking up the fight against the 2019 coronavirus, and on March 10, the team provided an update on their work so far, including open-source files on GitHub for researchers to use.
FAH uses volunteers’ personal computers when the computer is idle or isn’t doing any resource-intensive work to carry out research-related computations. By doing different tasks simultaneously on thousands of volunteer computers across the world, a concept known as distributed computing, the team hopes to significantly speed up the research process, even faster than if they were to use a supercomputer (more on this later).
Besides having incredible computational power, FAH is many magnitudes cheaper to operate than a supercomputer.
This isn’t FAH’s first rodeo. The project has been running since October 2000, and the Pande Lab has published 222 research papers (as of February 2020) as a direct result of the work done by FAH. As you would expect, hundreds of other researchers have also benefited from these papers. The project has contributed significantly to the understanding of:
- Neurological diseases like Alzheimer’s, Huntington’s, and Parkinson’s
- Infectious diseases like dengue, Zika, hepatitis C, and Ebola
- Breast cancer, kidney cancer, epigenetics, and the p53 protein
The primary means by which FAH contributes to these fields is by exploring a process called protein folding. Before explaining this term, here’s a recap from FAH on the fundamental role proteins play in our body:
Proteins are necklaces of amino acids, long-chain molecules. They are the basis of how biology gets things done. As enzymes, they are the driving force behind all of the biochemical reactions that make biology work. As structural elements, they are the main constituent of our bones, muscles, hair, skin and blood vessels. As antibodies, they recognize invading elements and allow the immune system to get rid of the unwanted invaders.
Proteins are the building blocks of the human body, but in order for proteins to perform any of these diverse functions, they first take on a particular shape. This process is known as protein folding.
When this does not happen as it should, it could lead to functional impairment and diseases like the ones mentioned above. Alzheimer’s disease, for example, is caused by misfolded proteins aggregating and forming clumps. Finding a treatment or developing a drug to cure a disease would require us to first understand how and why the underlying proteins misfold.
Understanding protein folding and computational biology
Accurate simulation of protein folding is considered the “holy grail of computational biology.” In the body, most proteins take milliseconds to seconds to fold completely. But for a computer to simulate all the atomic-level changes over this duration, it has to perform trillions of quadrillions of steps. If we were to use a single, high-end personal computer to simulate a millisecond-long protein fold, the simulation could take centuries to render. Even with Cray supercomputers, a millisecond-level simulation in 1998 took four months of CPU time. Instead of studying these slow-folding complex proteins, scientists were forced to study quickly folding proteins because these involved fewer steps and were less computationally intensive.
But with advances in processing power, millisecond-duration simulations are now a possibility. However, they are still computationally intensive. Hence, we have FAH, which relies on thousands of computers across the world to perform these simulations. FAH has the ability to simulate folding on a 1.5-millisecond timescale, which is a “thousand times longer than any previous atomic-level simulation,” according to FAH.
Why don’t scientists just use one giant supercomputer?
Folding@home is a supercomputer of sorts. Supercomputers are nothing but clusters of processors that are connected by fast networking. By using computers around the world, FAH is emulating this concept to some degree but without the fast networking aspect.
FAH is far more incredible in some ways. The performance of supercomputers is measured by the number of floating-point operations per second (FLOPS). Folding@home’s performance exceeded 100 petaFLOPS in July 2016. To put this in perspective, there is only one supercomputer in the world, the IBM Summit, that has a higher performance rate (150 petaFLOPS). In addition to this, the resources of a supercomputer are shared by multiple agencies and universities, which further reduces the available processing power for each research.
Besides having incredible computational power, FAH is many magnitudes cheaper to operate than a supercomputer. But most important to this case, according to the researchers in the Pande Lab, “Protein folding dynamics is statistical in nature, so a single long simulation from a supercomputer would not be sufficient to fully understand the folding process.”
Back to COVID-19
The latest coronavirus is similar to the SARS coronavirus from 2003. The infections resulting from both viruses occur in the lungs when a virus protein (spike protein) binds with a lung cell receptor protein (ACE2). The spike protein is depicted in red in the image above. An antibody prevents the spread of the infection by blocking the spike protein from binding with the receptor. To develop the antibody, scientists need to study the structures of the spike protein, the many shapes it takes, and how it binds to the receptor. Although we can use the knowledge gained while developing an antibody for SARS-CoV, the process still requires running lots of folding simulations unique to this strand of coronavirus. And that’s where FAH comes in.
Why don’t we use distributed computing for everything?
While Folding@home is one of the well-known distributed computing projects, there are many more out there that use idle computers to carry out various kinds of research in the fields of astronomy, chemistry, biology, climatology, mathematics, and physics. A leading initiative that is popular among researchers is the Berkeley Open Infrastructure for Network Computing (BOINC). Unlike FAH, which is purpose-built for protein folding, BOINC allows many other types of projects to use their software and take advantage of distributed computing. The World Community Grid is another initiative that aids research in health, poverty, and sustainability. FightAIDS@home and Outsmart Ebola Together are helping to save lives right now while SETI@home has been helping search for alien life for the last 20 years and ClimatePrediction.net is trying to forecast the weather 100 years into the future.
Whether you want to share your computer with FAH for the fight against the 2019 coronavirus or with one of the many other projects listed here and here, you now know how to contribute to some of the world’s most pressing challenges in your own small yet powerful way.