What started out as a competition between teams of students to build the most powerful computer evolved into an exploration at the frontier of molecular science. The students of Jon Bragdon's computer science class at St. Croix Regional Technical Center in Calais have constructed a supercomputer that participates in "protein folding" every evening as part of a global network of computers seeking cures for genetic disorders such as cancer and multiple sclerosis.
"     There are endless amounts of protein sequences that need to be calculated, and anybody can contribute," student Nathan Chambers explains.
     The initial task assigned to the students, who were divided into teams of four, was to research the best parts available to construct a supercomputer on a budget of $1,200. After assessing online reviews, students then searched for the best deal on the parts, ordered them and awaited their arrival. As the parts came in, they were assembled into the towers of each computer, and, once finished, a series of benchmarks was conducted.
     "This was all about maximizing performance on a budget with quantitative measurements in place," instructor Jon Bragdon says of the first phase of the project.
     Chambers' team constructed an impressive, water‑cooled supercomputer with eight computing cores running at 4.7 GHz and 32 gigabytes of memory running on a 1,000‑watt power supply. Another team, represented by student Antonio Andreucci, created the second top‑performing computer with four cores running at 3.5 GHz and 16 gigabytes of memory. In the benchmarks between the two computers, Andreucci's machine performed better, a difference the students attribute to the different brands of the processors. Chambers' team went with AMD while Andreucci's went with Intel.
     During the build process, Andreucci learned of a project being conducted by Stanford University called Folding@Home. According to the project's website, its goal is to assist scientists who are researching cures for various medical ailments by harnessing the unused computing power of computers all over the world. At any given moment, around 100,000 computers are networked for the task, each of them actively "folding" proteins to determine their function. When proteins are misfolded during the body's regular task of DNA replication, abnormalities such as cancer occur.
     "We know how to make proteins, but we'd like to know how these proteins behave," instructor Bragdon says. It isn't a problem that will be solved lightly. By way of illustrating the immensity of the task there are about 10 to the power of 80 particles in the universe. Proteins are chains composed of amino acids, and they may be of varying lengths. A single average protein length of 300 amino acids could consist of 4.6 x 10 to the power of 396 combinations.
      The task would be completely daunting if not for the power of the united network of computers. Bragdon explains that single supercomputers are typically extremely expensive when built on their own; however, when connected via the Internet as in the case of Folding@Home, the network created is one of the most powerful in the world.
      The work has proved beneficial from a scientific standpoint. Several proteins have been identified in their ability to affect Alzheimer's disease, cancer and viral infections. Hundreds of papers have been written utilizing the research from Folding@Home, and the amount of data produced is continually increasing with each core added to the network.
     Upon learning of the Folding@Home project, Andreucci connected his computer to the network to participate in the process. Since then, it has regularly folded approximately 10 proteins a night, with up to 40 completed each weekend. After Andreucci's semester at SCRTC is over, the computer parts will be repurposed for the next class and whatever project they devise. Andreucci and his classmates regard the experiment as a valuable learning experience.
     While the effort to cure fatal and costly medical conditions may be admirable, Bragdon admits that there are other factors to consider when discussing the decoding of the building blocks of life. "It has huge ethical implications," Bragdon says. "That's true for all technology. It's all in the framework of the user."