Banner by Georgia Barry

Become A Protein Pro With Folding@home

By Sarah Lindley

Lights off. Fan on. Clean sheets. You have dozed off, but your computer, plugged in and sitting on your desk, is as awake as it can be. It is crunching numbers, delivering and receiving calculations, and running simulations, all while you dream. The computer is silently sending mathematical outputs to biology researchers, who use these solutions to change the world one protein at a time. You are helping science progress by providing the materials to solve some of the most complex biological problems known to man, all while you are asleep.

Why are researchers so interested in proteins? These large, complex molecules make up your skin, hair, muscles, and most of your body tissue. They do the work that keeps you alive, like carrying oxygen through your blood. The instructions for building proteins are encoded in your DNA, but once the base materials are put together, most proteins then have the task of self-assembly, folding themselves into the correct shape in order to function. Sometimes, however, they do not do so correctly, which can lead to problems such as Alzheimer’s, Parkinson’s, and other muscular degenerative and neurodegenerative diseases. Proteins are crucial to life, but if they are folded incorrectly, they can produce devastating consequences.

Unfortunately, creating drugs to treat diseases caused by misfolded proteins is easier said than done. Researchers generally do not know what most proteins look like in three-dimensional space, since the details are too small to be seen through a microscope, and it is challenging to target misfolded proteins when we do not know exactly where the mistakes are. To tackle this problem, researchers have turned to computers to run calculations that simulate protein folding. However, completing most of these computations (which are extremely complex) would take a single computer centuries. Luckily, Folding@home is here to help.

Founded by Dr. Vijay Pande at Stanford University in 2000, Folding@home is a program anyone can download to help simulate protein folding and automatically send the results to researchers. It uses your device’s computing power to calculate trajectories of every individual atom in a protein, each of which is operating on a randomly assigned, temperature-dependent velocity. The program then runs thousands of simulations resulting from these calculations, most of which do not produce valuable results, but some of which do lead to a protein successfully folding. With data pooled from all of its users, Folding@home clumps together simulations that produce similar results to provide insight on experimental procedures and build maps of the proteins.

One of the leading proponents of the program is Dr. Greg Bowman from Washington University in St. Louis, who uses it often in his lab. Bowman explains that the goal of Folding@home is to build a “satellite” image of proteins. When your computer sends back its data, you are expanding the information available and helping to build that composite, similar to how a collection of many, many aircraft satellite images were brought together to create Google Earth.

This model, in which data is sent between servers and computers that complete part of a complex process, is known as distributed computing. To understand distributed computing, imagine if you had a 10-chapter test and there were 10 of you who could each study a different chapter at the same time. You would be studying for much less time and each version of you would only have to do 10% of the total work. Distributed computing systems likewise allow thousands of computers to contribute a small piece of the protein puzzle simultaneously, instead of one computer taking years to do all the work.

Here is how it works on your computer: whenever you are not using it and you have the program enabled, the Folding@home server requests that your computer completes specific computations that are part of a larger task. One task is assigned to many computers, so that no single computer spends an unreasonable amount of time or computing power finishing its portion. Because this all occurs simultaneously, it significantly reduces the length of time these calculations take compared to even a supercomputer.

Folding@home is considered a supercomputer itself, and after it started focusing its research on proteins related to SARS-CoV-2, the virus that causes COVID-19, it officially had “more raw computing power than the world’s largest 500 traditional supercomputers combined.” Four months into the transition, the amount of Folding@home users had increased a hundredfold, with companies like Amazon Web Services, Microsoft, Cisco, and more donating some of the computing power from their cloud systems to the effort. Folding@home also received a National Science Foundation grant called a Rapid Response Research (RAPID) award to study the virus. RAPID awards go to research projects that are emergent or time-sensitive, and which are expected to have a high impact on public well-being; past grants have gone to researchers studying the Ebola and Zika viruses, as well as responses to major disasters such as hurricanes and oil spills. To further expedite the process of SARS-CoV-2 research, Folding@home promised to share all its related data in public databases and to publish its findings in free, pre-print journals, such as bioRxiv.

While Folding@home has been instrumental for countless research projects related to proteins over the last two decades, it also comes with potential concerns of its own, including energy use. By having your computer run while idle, it has the potential to use an excessive amount of energy, raise electricity bills, and contribute to climate change. Folding@home’s operators have acknowledged this issue and stated that they pursue projects that will have a “direct scientific impact and measurable positive influence on the human condition,” not projects about which researchers are merely curious. Since the goals of the projects being completed are shared with the users, the operators recognize that the users may stop contributing if they disapprove of a project. Each contribution is valuable, so the operators strive to select worthwhile projects; that way, users never end up having to express their disagreement by uninstalling Folding@home. As a means of combating excessive electricity usage on a personal level, users are able to decide how much power they want to contribute to a project, and can stop and start the program at any time. According to Folding@home’s FAQ page, running the program 24 hours a day can increase electricity costs by about $0.36 a day, which could be conserved by turning off lights and computer monitors, though this varies depending on location and how powerful the computer’s central processing unit (CPU) is.

Another concern that comes with connecting your personal computer to a centralized server is data security, but on this front, Folding@home seems to have a clean record. The program puts a unique signature made of 2048 bits (zeros and ones) on all data going to and from personal computers, and if something has interfered and the signatures do not match, the data is discarded. As long as the software is downloaded from the correct location—the Folding@home website—there seems to be no reason for alarm. Folding@home has not yet suffered any data leaks, and the only files it reads and writes on your computer are those associated with the program.

If you want to get started with Folding@home on your own computer, you can download the program from foldingathome.org. Once it is set up, you will be able to specify how much power you want the program to use, and if you want it to be running all the time or only while your computer is idle. You can choose which disease you want your computer to contribute research to, and if you would like to get your friends and family involved, you can create a team and earn points together. Folding@home will give you information about the work unit or task your computer is completing and what research project it is contributing to. For example, my computer is currently contributing to Project 16978, a project managed by Temple University that is researching halogen bond donors and acceptors in relation to cancer.

In my experience, Folding@home is easy to operate and user-friendly; after all, it will be the computer doing the math, not you. Think about it: you, too, could be making an impact on the scientific community and medical research, without really trying. Just try not to fold under the pressure while trying to decide.