Home
Organization
Facilities
Opportunities
Research
Research Tools
Publications
Highlights
In the News
Search
ORNL Protein Fold Prediction Program Succeeds in CASP3 Contest (June 1998)The Computational Biosciences Section in Life Sciences Division at Oak Ridge National Laboratory has developed a new computer based system for protein tertiary fold prediction. The program predicts the approximate tertiary structure of a protein sequence based on the recognition of structures or partial structures in a large database of known protein folds that are potentially similar to an unknown. The recognition of similar structures or partial structures is generally achieved by finding a "good" alignment between known protein structures and the query protein sequence based on a recently derived set of hybrid potential energy functions. A unique feature of this program is that it guarantees to find an alignment that is globally optimal, and does so very efficiently (with polynomial-time running efficiency). This represents a new breakthrough in the area of fold recognition-based protein structure prediction methods and is being used successfully in the current CASP3 protein structure prediction contest. The most recent predicted structure, an E. coli pyrophosphokinase, HPPK, matched the just completed crystallographic structure with good fidelity. The core algorithm is described in an invited paper, entitled "An efficient computational method for globally optimal threading", in press in the Journal of Computational Biology, with several more papers containing detailed descriptions of energy functions and system performance expected later this year. A number of research groups from organizations such as NIH, LBL, Amgen, and Boston University have expressed interest in utilizing the system in their research and in collaborative development related to the project.Contact: E. C. Uberbacher, 574-6134 or ube@ornl.gov Funding Source: DOE-OBER |
Oak Ridge National Laboratory is operated by UT-Battelle, LLC,
under contract DE-AC05-00OR22725 for the U.S. Department of Energy.