ABSTRACT
Proteins are biopolymers that present the particularity to adopt a specific tridimensional structure dictated by their amino acid sequence. As they are generally only active in their folded form, it is important to get information about their 3D structure, whether by in vitro or in silico means. In silico approaches to protein folding and misfolding are becoming increasingly powerful tools for rationalizing, simulating, and understanding data obtained in vitro, and should lead eventually to rational protein design. The efficiency of protein structure prediction thus depends strongly on the availability of similar sequences or folds: the more similar the target is to proteins of known structure, the easier is the prediction of its structure. Number of proteins are likely to adopt alternative, nonnative, structures that can be at the basis of conformational diseases.
