It is well known that protein function is directly linked to the three-dimensional protein structure. The eventual goal of structural genomics to understand the correlation between protein structure and enzyme function. The rapid increase in number of known protein structures in the Protein Data Bank (PDB) has given rise to the need for algorithms and tools that illuminate the linkage between protein structure and function. Detection of key functionally active amino acids is necessary for protein classification, evolutionary study and drug design. A novel approach to prediction functional sites within proteins involves the application of machine learning algorithms such as neural nets and random forests to the topological space created through the Delaunay tessellation of a proteins Calpha backbone. This study focuses on the development methods for site identification proteins through Delaunay tessellation of known protein structures in hopes of further defining the structure/function correlation.
展开▼
