The Fred DeMatteis School of Engineering and Applied Science at Hofstra University is lending the extensive resources of its Big Data Laboratory to a global project studying the protein malformations that can lead to coronavirus and other conditions such as Alzheimer’s disease, cystic fibrosis, and cancer.
The Folding@home initiative supports biomedical research at academic labs around the world through donated computing power. The long-running, crowdsourced project runs large-scale, complex protein folding simulations through software that is downloaded on multiple machines to improve efficiency.
Hofstra University is currently in the top one percent of 253,000 teams who are providing computational power to this worldwide consortium, which began at Stanford University and now includes labs at Memorial Sloan Kettering Cancer Center, Temple University, University of Illinois, and Stockholm University, among others. Corporate collaborators have included Intel, Microsoft, and Google.
“Computers are idle as the COVID-19 pandemic has forced the shutdown of businesses and institutions around the world, and we can use this time to help carry out the demanding, time-consuming computations that are required in virus research,” said Dr. Edward Currie, associate professor of electrical engineering. “Hofstra’s sophisticated computer system is currently almost fully dedicated to this effort.”
Proteins are essential building blocks of the human body. On a molecular level, a protein structure “folds” on itself, an ongoing process that determines its particular function in keeping the body healthy. Disruptions in this process are implicated in many diseases – for example, COVID-19 has viral proteins that reproduce and work to suppress the immune system. Researchers use complex computer simulations of molecular interactions to study protein dynamics and identify these folding errors, with the hope of developing new and improved therapies to target them.
“Since the start of the COVID-19 pandemic, Folding@home has seen a significant surge in downloads, a clear indication that people around the world are concerned about doing their part to help researchers find a remedy to this virus,” said Dr. Sina Rabbany, dean of the DeMatteis School. “We are thrilled that Hofstra can contribute to this important initiative at a time when we are facing one of the worst global public health crises the modern world has seen.”
The effort at Hofstra is led by Dr. Currie and Alex Rosenberg, the DeMatteis School’s systems administrator who oversees the Big Data Laboratory for the Department of Computer Science. More than 130 machines, housed in two buildings, with a combined equivalent of 900 threads are running the Folding@home program. (For perspective, a typical desktop computer has a quad-core processor that can run four or eight threads concurrently, so 900 threads is comparable to the computing power of as many as 225 computers.) Altogether, Folding@home’s global network now comprises an “exaflop” of computing power: 1,000,000,000,000,000,000 operations per second.
“It is part of our mission to help solve problems for the betterment of human society,” said Dr. Krishnan Pillaipakkamnatt, chair of the Department of Computer Science. “We are fortunate to have powerful resources, strong infrastructure, and dedicated faculty and staff to contribute in a meaningful way.”
The Folding@home project is based at the Bowman Lab at Washington State University and relies on individuals (“citizen scientists”) and teams who volunteer to run the free application with unused power on their home computers.