The influence of tertiary structure on secondary structure prediction Accessibility versus predictability for β-structure

AbstractThe secondary structure prediction algorithm of Garnier et al. [(1978) J. Mol. Biol. 120, 97-120] has been used for 16 proteins whose structures are dominated by β-sheet. Comparisons of the predicted structures with those defined by the algorithm of Kabch and Sander [(1983) Biopolymers 22, 2577-2637] shows that for β-sheet residues, the quality of prediction falls markedly with increasing residue accessibility. 2 sub-classes of β-residues have been distinguished on the basis of hydrogen bonding patterns, and the distribution of amino acid types within each sub-class found to be quite different. Accordingly, Chou and Fasman Pβ-type parameters for these previously indistinguished states have been derived