Project Highlights
Blue Gene/L Helps Improve Quality of Protein Structure Predictions
 |
One of the key challenges in computational biology is the ability to predict the three-dimensional structure of a protein from its sequence. De-novo or template-based modeling approaches are adopted, depending on whether the protein of interest is similar to a protein of known structure that can be used as a template for building a model. It is known that for most proteins the native state lies at the bottom of a free-energy landscape. The protein structure prediction problem involves varying the degrees of freedom of the protein in a constrained manner until it reaches its native state or close to it.
For our INCITE project, we have used the Rosetta protein modeling software for predicting protein structures, using both de-novo and template-based techniques. In the Rosetta protein structure prediction protocols, a very large number of independent trajectories of folding simulations are started. At the end of the simulation, the final conformation of trajectory rests in a local minima. A small fraction of these conformations are substantially lower in energy and are predicted to be close to the native state. Given that the size of the conformation space to be sampled is huge, a very large number of models have to be generated in order to find a small subset close to the native conformation.
Aided by the Blue Gene/L supercomputer, we are able to sample a substantially larger pool of conformations. This is about two to three orders of magnitude greater than was previously possible. We have sampled closer to near-native conformations, resulting in an overall improvement in the quality of our predictions.
