Hierarchical energy-based approach to protein-structure prediction: Blind-test evaluation with CASP3 targets

A hierarchical approach based exclusively on finding the global minimum of an appropriate potential energy function, without the aid of secondary stru cture prediction, multiple-sequence alignment, or threading, is proposed. T he procedure starts from an extensive search of the conformational space of a protein, using our recently developed united-residue off-lattice UNRES f orce field and the conformational space annealing (CSA) method. The structu res obtained in the search are clustered into families and ranked according to their UNRES energy Structures within a preassigned energy cutoff are gr adually converted into an all-atom representation, followed by a limited co nformational search at the all-atom level, using the electrostatically driv en Monte Carlo (EDMC) method and the ECEPP/3 force field including hydratio n. The approach was tested (in the CASP3 experiment) in blind predictions o n seven targets, five of which were globular proteins with sizes ranging fr om 89 to 140 amino acid residues. Comparison of the computed lowest-energy structures, with the experimental structures, made available after the pred ictions were submitted, shows that large fragments (similar to 60 residues, representing 45-80% of the proteins) of those five globular proteins were predicted with the root mean square deviations (RMSDs) ranging from 4 to 7 Angstrom for the C-alpha atoms, with correct secondary structure and topolo gy. These results constitute an important step toward the prediction of pro tein structure based solely on global optimization of a potential energy fu nction for a given amino acid sequence. (C) 2000 John Wiley & Sons, Inc.