Item

Abstract

IT is generally accepted that the action of a protein cannot be understood until its three-dimensional structure is known. At present, X-ray analysis of protein crystals is the only method of obtaining such structural information. It is to be feared, however, that many important proteins will never give suitable crystals so that one is obliged to consider other approaches to structure elucidation. Renaturation experiments indicated1–4 that the three-dimensional structure of many if not all proteins is a unique function of their amino acid sequence. Therefore, in principle one should be able to determine these structures by using only the information contained in the sequence. A first step in this direction is the prediction of secondary structures (α helices, β pleated sheets, β bends) in globular proteins from amino acid sequences. Several prediction schemes have been devised to this end5–23. It is the aim of this paper to demonstrate directly the current standing of such methods.