THE CHARACTERISTICS OF COMMON SEQUENCE PEPTIDES IN PROTEINS. Luhua Lai, Dawei Lin, Youqi Tang, Institute of Physical Chemistry, Peking University, Beijing 100871, China

Characteristics of sequentially identical short peptides in a protein 3 dimensional representative data set and their relationship with protein folding were studied. Sequentially identical short peptides have the same intrinsic properties. They are ideal candidates for studying how short peptides conformation are influenced by their surrounding structural environment. We have carried out an exhaustive search on sequentially identical peptides in a protein structure representative set generated from the Brookhaven Protein Data Bank. The relationships among their sequence, secondary structure, protein fold class and short peptide structure adaptability in protein were analyzed. Six examples were found in the same protein fold class but having totally different structures. The proteins they are in have sequence identity less than 50%. This result implies that the conformation of some short peptides are influenced by more detailed structural environment than protein fold class, so that protein fold class dependent secondary structure prediction algorithm will still encounter the structural plasticity dilemma. From the sequence and structure analysis, we found that although some of the sequentially identical peptide pairs take different structures, most of the sequentially identical sequence peptide pairs have similar structure. A positive correlation was found between the accuracy of secondary structure prediction and the structure conservation of common sequence short peptides. The common sequence peptide pairs which preserve their structure in unrelated protein and different local structural environment have been proposed to be in the folding nucleation site or folding initiation site. One of them has found experimental evidence. The results of this study give helpful clues to protein secondary structure prediction and folding study.