A volumetric score function for computational protein structure modeling and protein structure validation

We present a novel method, RosettaHoles2, for visual and quantitative assessment of underpacking in the protein core. For visualization, an explicit representation of void volume is provided in the form of clusters of overlapping spheres that fill interatomic space in the molecule. For quantitative analysis, two volumetric score functions are presented for protein structure refinement and validation. The refinement score is parameterized to maximize the gap between computationally generated and decoy structures; it is differentiable and can be easily incorporated into existing molecular forcefields. The parameters of the refinement score provide insight into components missing in current forcefields. The validation score provides an upper bound on the X-Ray resolution of protein data bank structures; a crystal structure with acceptable packing properties will have a validation score as good as or better than its resolution. RosettaHoles2 provides excellent discrimination between real and computationally generated structures, is predictive of incorrect regions in models, identifies problematic structures in the Protein Data Bank, and promises to be a useful validation tool for newly solved experimental structures.