Automatic construction of 3D structural motifs for protein function prediction

Structural genomics initiatives are on the verge of generating a vast number of protein structures. The biological roles for many of these proteins are still unknown, and high-throughput methods for determining their function are necessary. Understanding the function of these proteins will have profound impact in drug development and protein engineering. Current methods for protein function prediction on structures require manual creation of structural motifs. Thus only few structural motifs are available. The lack of structural motifs limits the use of these methods for function prediction at a structural-genomics scale. To overcome this limitation, we describe a method for automatically creating a library of three dimensional structural motifs. Automatically generating a library of structural motifs can be used for structural-genomic scale function prediction on protein structures.