Non linear signal analysis methods in the elucidation of protein sequence/structure relationships

We tried to present the general meaning and scope of the application of signal analysis methods to protein sequence-structure relationships. In our opinion, the dominant character of this nascent field concerns the number of the contributing disciplines, both in terms of methodological (the data analysis techniques come directly from engineering, applied mathematics, computational physics) and theoretical influence (the basic assumptions informing the various approaches coming from molecular biology, evolutionary genetics, physical chemistry, biochemistry). As eloquently discussed in the Laughlin et al. paper entitled “The Middle Way” (P. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 32) the frontiers of science are rapidly shifting from the investigation of the basic bricks of matter to the elucidation of mesoscopic principles of organization. The study of proteins is perhaps the most typical “mesoscopic” investigation field with its mixing of basic physical laws, empirical results, and qualitative descriptions.4 In this review we describe a particular approach that adopts an empirical style typical of medicinal chemistry, letting general principles emerge from the practical solution of local cases.