HYPO: Computational Prediction of (Hy)drogen (Po)sitions in Protein Structures from X-Ray and Neutron Crystallographic Data

Abstract

X-ray crystallography has become the standard means for gathering experimental structural data from proteins. However, there are problems inherent with X-ray crystallography, most especially with the explicit location of hydrogen atoms. This poses a problem, in that many critical functions carried out by protein active sites rely on fine positioning of key hydrogen atoms. Another experimental technique is neutron crystallography, which is especially good at determining hydrogen positions with a high level of accuracy, but which carries extreme difficulty. Our project aims to allow for a more systematic, accurate, and quantitative methodology for computationally locating hydrogen atom positions within proteins, using data gathered from high resolution X-ray and neutron crystallographic structures. Use of these methods will make crystallographers less dependent upon ultra-high resolution crystal structures and experimentally challenging neutron data when determining locations of critical hydrogen atoms within a protein structure.