Deep learning model predicts water interaction sites on the surface of proteins using limited-resolution data

Learning to Identify Physiological and Adventitious Metal-Binding Sites in the Three-Dimensional Structures of Proteins by Following the Hints of a Deep Neural Network

Thirty-eight percent of protein structures in the Protein Data Bank contain at least one metal ion. However, not all these metal sites are biologically relevant. Cations present as impurities during sample preparation or in the crystallization buffer can cause the formation of protein-metal complexes that do not exist in vivo. We implemented a deep learning approach to build a classifier able to distinguish between physiological and adventitious zinc-binding sites in the 3D structures of metalloproteins. We trained the classifier using manually annotated sites extracted from the MetalPDB database. Using a 10-fold cross validation procedure, the classifier achieved an accuracy of about 90%. The same neural classifier could predict the physiological relevance of non-heme mononuclear iron sites with an accuracy of nearly 80%, suggesting that the rules learned on zinc sites have general relevance. By quantifying the relative importance of the features describing the input zinc sites from the network perspective and by analyzing the characteristics of the MetalPDB datasets, we inferred some common principles. Physiological sites present a low solvent accessibility of the aminoacids forming coordination bonds with the metal ion (the metal ligands), a relatively large number of residues in the metal environment (≥20), and a distinct pattern of conservation of Cys and His residues in the site. Adventitious sites, on the other hand, tend to have a low number of donor atoms from the polypeptide chain (often one or two). These observations support the evaluation of the physiological relevance of novel metal-binding sites in protein structures.

Structural genomics and the Protein Data Bank

The field of Structural Genomics arose over the last 3 decades to address a large and rapidly growing divergence between microbial genomic, functional, and structural data. Several international programs took advantage of the vast genomic sequence information and evaluated the feasibility of structure determination for expanded and newly discovered protein families. As a consequence, structural genomics has developed structure-determination pipelines and applied them to a wide range of novel, uncharacterized proteins, often from “microbial dark matter,” and later to proteins from human pathogens. Advances were especially needed in protein production and rapid de novo structure solution. The experimental three-dimensional models were promptly made public, facilitating structure determination of other members of the family and helping to understand their molecular and biochemical functions. Improvements in experimental methods and databases resulted in fast progress in molecular and structural biology. The Protein Data Bank structure repository played a central role in the coordination of structural genomics efforts and the structural biology community as a whole. It facilitated development of standards and validation tools essential for maintaining high quality of deposited structural data.