‘Meta’ Approaches to Protein Structure Prediction

The computational assignment of three-dimensional structures to newly determined protein sequences is becoming an increasingly important element in experimental structure determination and in structural genomics (Fischer et al. 2001a). In particular, fold-recognition methods aim to predict approximate three-dimensional (3D) models for proteins bearing no evident sequence similarity to any protein of known structure (see the review by Cymerman et al., this Vol.). The assignment is carried out by searching a library of known structures (usually obtained from the Protein Data Bank) for a compatible fold. A variety of fold-recognition methods has been published, both structure-dependent (i.e.threading) (Sippl and Weitckus 1992; Godzik et al. 1992; Jones et al. 1992; Ouzounis et al. 1993; Bryant and Lawrence 1993; Rost 1995; Alexandrov et al. 1996; Di Francesco et al. 1997; Fischer 2000; Kelley et al. 2000; Shi et al. 2001) and sequence-only dependent (Karplus et al. 1998; Rychlewski et al. 2000). The state-of-the-art in the field of fold recognition is currently to combine the evolutionary information available from multiple sequence alignments for the target and the template (to detect remote homology between protein families) and the structural information from the template (to detect similarities of folds of compared proteins regardless of their evolutionary relationship, i.e. analogs and homologues as well).