"Protein folding is a big problem, there are a large number of proteins and a lot of possible shapes/fold," explained NYU scientist Bonneau, a new faculty member at NYU's Center for Comparative Functional Genomics, with a joint appointment in Biology and Computer Sciences. "In spite of the difficulty, it is an important problem, at the heart of deciphering genomes. The shear amount of compute power needed to carry out this project makes the use of grid computing essential."
It is for this reason that scientists at NYU have teamed up with IBM. The World Community Grid (wcgrid.org) aims to create the world's largest public computing grid to undertake projects that benefit humanity. IBM has developed the technical infrastructure that serves as the grid's foundation for scientific research.
The first and second phases of the NYU research are part of the Human Proteome Folding project (HPF), which combines the power the idle cycles on millions of computers (which we call the grid) to help scientists understand how human proteins fold, the shapes they take on after folding. As computers try millions of ways to fold the chains, they attempt to fold the protein in the way it actually folds in the human body (accurately predict the structure). The best shapes/3D-structures identified for each protein are returned to the scientists for further study and public release. Knowing the shapes of proteins will help researchers understand how proteins perform their functions in vivo (in the cell) and the roles of proteins in diseases. With a greater understanding of protein structure, scientists can learn more about the biological systems that underlie most human activity (biomedical, agricultural, environments). In the end, this work is enabled by the people, around the world, who have volunteered their idle cycles by downloading the grid client (wcgrid.org).
In the first phase, NYU biologists, headed by Professor Richard Bonneau, obtained structure predictions for more than 150 genomes. For more on the Bonneau laboratory's findings, go to http://www.cs.nyu.edu/~bonneau/Struct-pred.html. In this first phase, the NYU team employed "Rosetta," a computer program used in predicting de novo protein structure--"de novo" is the modeling of proteins when there is no "real world" structure on which to base predictions.
"With the first phase we aimed to get protein function by predicting the shape of many protein structures," explained Bonneau. "With the second phase, we will increase the resolution of a select subset of human proteins (attempt to determine the structure with respect to all atoms in the molecule). This phase also include a large test set and will thus serve to improve our understanding of protein structure prediction and advance the state of the art in protein structure prediction."
The NYU researchers, working with researchers studying new methods for early detection of cancer at Seattle's Institute for Systems Biology, will focus on cancer biomarkers--proteins expressed during the early stages of several cancers. They will also focus on proteins involved in host-parasite interactions that are key to our understanding of malaria. They will use a different mode of the Rosetta program to generate higher resolution structures, thereby refining predictions from the first phase with more accurate but also much more computationally demanding methods.