The goal of this project is to develop a structure description and associated sequence-to-model structure alignment scoring (evaluation schema) applicable to general inverse protein folding methodologies. The approach is motivated by the recognition that there are currently no practical means of predicting a protein's structure directly from its sequence. The project's approach is based on a domain fold structural description. Pseudo potential functions are derived from the observed physical and/or conserved residue interactions parameterized by the natural tetrahedral geometry. The information theory methods are applied to optimize the most predictive pseudo potential function. This project has the potential of providing a major step toward making the threading approach to protein structure prediction and modeling a more viable tool. It will provide a better understanding of the underling mathematical basis for developing Gibbsian scoring schema for any sequence-to-structure alignment method and will aid in the development of more efficient lower bound calculation for the threading Brand-and-Bound optimal alignment algorithm.