Improved recognition of native-like protein structures using a combinationof sequence-dependent and sequence-independent features of proteins

me describe the development of a scoring function based on the decompositio n P(structure\sequence) proportional to P(sequence\structure) *P(structure) , which outperforms previous scoring functions in correctly identifying nat ive-like protein structures in large ensembles of compact decoys. The first term captures sequence-dependent features of protein structures, such as t he burial of hydrophobic residues in the core, the second term, universal s equence-independent features, such as the assembly of beta-strands into bet a-sheets. The efficacies of a wide variety of sequence-dependent and sequen ce-independent features of protein structures for recognizing native-like s tructures were systematically evaluated using ensembles of similar to 30,00 0 compact conformations with fixed secondary structure for each of 17 small protein domains. The best results were obtained using a core scoring funct ion with P(sequence\structure) parameterized similarly to our previous work (Simons et al., J Mol Biol 1997;268:209-225] and P(structure) focused on s econdary structure packing preferences; while several additional features h ad some discriminately power on their own, they did not provide any additio nal discriminatory power when combined with the core scoring function. Our results, on both the training set and the independent decoy set of Park and Levitt (J Mol Biol 1996;258:367-392), suggest that this scoring function s hould contribute to the prediction of tertiary structure from knowledge of sequence and secondary structure. (C) 1999 Wiley-Liss, Inc.