Evaluating the potential of using fold-recognition models for molecular replacement.

Here, the proposal is investigated that protein tertiary structure prediction methods and threading methods in particular might be applied to the problem of solving a protein structure by X-ray crystallography, thus reducing the need for the more traditional experimental intensity methods of data phasing, such as heavy-metal isomorphous replacement and anomalous scattering methods, and without reference to a very closely related protein of known structure. If this kind of approach were to become successful and reliable, this would represent a significant advance in protein structure determination, offering an easy and accessible method for the initial data phasing for proteins' crystal structures, utilizing the vast amount of structural data, deposited in the Brookhaven PDB, that has been accumulated over the past 30 years of crystallographic structural studies. In the light of the ongoing structural genomics initiatives, the successful development of this kind of approach would be of enormous benefit.