The prediction of protein structure and function remains a difficult problem despite significant advances in this field. Current methods involve comparison of unknown proteins with proteins whose structures and functions are already known. These techniques require that proteins of similar structure and function have similar sequences but this is not always the case. This book examines how spectral techniques may be used to predict protein structure and function based on amino acid sequences alone, without relying on the comparative methods currently in use. Two different approaches are attempted. The spectral covariance is used with cluster analysis and the spectral envelope with classification and regression trees (CART). The results obtained open many avenues for further research and investigation.

Данное издание не является оригинальным. Книга печатается по технологии принт-он-деманд после получения заказа.