PhD defense will take place on Wednesday June 6, 2012 in room 3A75 Swearingen at 10 am. Matthew Fawcett Protein folding is the one of the most important topics of research in Biochemistry. The understanding of how a protein folds into its three dimensional shape from its amino acid sequence is of extreme importance in medicine today because proteins play important roles in diseases such as sickle cell anemia, Alzheimer’s disease, and Parkinson’s disease to name a few. Protein structures were traditionally determined by x-ray crystallography and more recently with nmr spectroscopy. However these methods are expensive and take a long time to perform. Recently, computational methods such as: ab initio, homology, and threading methods, have gained recogni- tion. The thermodynamics hypothesis of proteins states that a protein’s native structure is determined by its amino acid sequence, and resides in the lowest energy state. Ab initio methods provide the means to calculate the energy of proteins using an all atom forcefield. Traditional forcefields, like ones used in CHARMM and Xplor-NIH, use energy terms that are in two categories: bonded and non-bonded terms. However, there are problems with traditional forcefields in that, the energy landscapes are too complicated with too many variables. Also, the energy landscape does not exhibit a funneling effect to guide the protein down to the native conformation. Finally there is a notable lack of open source software to provide the community of Computer Scientists, Mathematicians, and Engineers with an opportunity to be engaged in this line of investigation and integrate their expertise. Here we propose addressing the above problems through the development of the software package Semi Classical Open-Source Protein Energy (SCOPE). SCOPE is an open-source C++ program that will reconstruct a protein within the rotamer space. Representation of a protein structure in the rotamer space reduces the degrees of freedom of the protein structure conformational space, allowing SCOPE forcefield to only have to calculate the non-bonded energy terms. The energy landscape will be simplified due to the reduction in the energy terms of the forcefield. Also, SCOPE will extend the traditional forcefield by calculating a hydrogen bond term along with the number of consecutive hydrogen bonds as they appear in a protein. The use of the extra hydrogen bond term is likely to create a funneling effect in the protein’s energy landscape to its native conformation. The program is also open-source which will allow the community of Computer Scientists, Mathematicians, and Engineers a chance to add to the solution without investing significant time and effort in becoming familiar with the field of biophysics. SCOPE’s energy profiles are incorporated into an artificial neural network(ANN) based approach to predict the backbone root mean square deviation(BB RMSD) of a given protein structure to that of the native structure based on its potential energy alone. This shows a remarkable correlation between the energy profiles and BB RMSD of a protein. The correlation was as high 0.99 in several experiments. The ANN can be used to help refine a protein from around 7.0 Å to approximately 2.0 Å.