3D Protein Folds: Homologs Against Errors—a Simple Estimate Based on the Random Energy Model

One cannot predict the 3D structure of a protein directly from its sequence because of errors in the energy estimates. However, using a set of homologs (proteins with nearly identical 3D structures despite numerous amino acid mutations in their chains) it is possible to average the fold energies over the homologs (this diminishes energy errors) and to predict the common 3D fold of these proteins from the set of their amino acid sequences.