Prediction of beta-turns from amino acid sequences using the residue-coupled model

Beta-hairpin prediction with quadratic discriminant analysis using diversity measure

On the basis of the features of protein sequential pattern, we used the method of increment of diversity combined with quadratic discriminant analysis (IDQD) to predict beta-hairpins motifs in protein sequences. Three rules are used to extract the raw beta-beta motifs sequential patterns for fixed-length. Amino acid basic compositions, dipeptide components, and amino acid composition distribution are combined to represent the compositional features. Eighteen feature variables on a sequential pattern to be predicted are defined in terms of ID. They are integrated in a single formal framework given by IDQD. The method is trained and tested on ArchDB40 dataset containing 3088 proteins. The overall accuracy of prediction and Matthew's correlation coefficient for the independent testing dataset are 81.7% and 0.60, respectively. In addition, a higher accuracy of 84.5% and Matthew's correlation coefficient of 0.68 for the independent testing dataset are obtained on a dataset previously used by Kumar et al. (Nucleic Acids Res 2005, 33, 154), which contains 2088 proteins. For a fair assessment of our method, the performance is also evaluated on all 63 proteins used in CASP6. The overall accuracy of prediction is 74.2% for the independent testing dataset.

Conformational studies of peptides containing cis-3-hydroxy-D-proline

Conformational analysis of peptides containing cis-3-hydroxy-d-proline (d-cis-3-Hyp) by NMR studies revealed that the 3-hydroxyl group in this amino acid plays a significant role in the overall three-dimensional structures of the peptides. When the d-cis-3-Hyp had its 3-hydroxyl group protected as the benzyl (Bn) ether, the peptide displayed a beta-hairpin structure in both CDCl(3) and DMSO-d(6). Even after the removal of the Bn group, the resulting deprotected compound retained the same structure as in the protected version in CDCl(3). However, in polar solvent DMSO-d(6), the C-terminal strand of the hydroxyl-deprotected peptide flipped to the side of the hydroxyl group, breaking the hairpin to form a pseudo beta-turn-like nine-membered ring structure involving an intramolecular hydrogen bond between LeuNH --> HypC3-OH.

Prediction of gamma-turns from amino acid sequences

We predicted gamma-turns from amino acid sequences using the first-order Markov chain theory and enlarged representative data sets corresponding to protein chains selected from the Protein Data Bank (PDB). The following data sets were used for training and deriving the probability values: (1) an initial data set containing 315 protein chains comprising 904 gamma-turns and (2) a later data set in order to include new entries in the PDB, containing 434 protein chains and comprising 1053 gamma-turns. By excluding 93 protein chains that were common to these two training data sets, we generated two mutually exclusive data sets containing 222 and 341 protein chains for testing our predictions. Applying amino acid probability values derived from training data sets on to testing data sets yielded overall prediction accuracies in the range 54-57%. We recommend the use of probability values derived from the data set comprising 315 protein chains that represents more gamma-turns and also provides better predictions.