Computational mutagenesis by using AUTO-MUTE 2.0 to examine X-ray crystallography structures for p53 and the impact of its resolution on predictions

ABSTRACT The mutation of P53 is found in 50% of all human cancers, affecting the protein structure conformation, which further impacts its function. Identifying the appropriate protein structure for future research has been challenging. Many X-ray crystallography structures exist in the Protein Data Bank (PDB) for a given protein. We compared two proteins with different resolutions to determine whether the differences between the two structures were statistically significant. The present study used mutagenesis to study two structures of the protein P53 with various resolutions to assess how the conformation of the protein changes upon mutation. We used an AUTO-MUTE 2.0, which is a computational geometry method based on the Delaunay tessellation for predicting protein structures. The current study compares the prediction results obtained from different predictive methods. We used logistic regression models and corresponding receiver operating characteristic (ROC) curves to model the differences between structures using the variables available at PDB. We conclude from this study that x-ray crystallography is a sensitive and advanced technique, and the resulting proteins at different resolutions are similar.