PDBspheres - a method for finding 3D similarities in local regions in proteins

We present a structure-based method for finding and evaluating structural similarities in protein regions relevant to ligand binding. PDBspheres comprises an exhaustive library of protein structure regions (spheres) adjacent to complexed ligands derived from the Protein Data Bank (PDB), along with methods to find and evaluate structural matches between a protein of interest and spheres in the library. Currently, PDBspheres library contains more than 2 million spheres, organized to facilitate searches by sequence and/or structure similarity of protein-ligand binding sites or interfaces between interacting molecules. PDBspheres uses the LGA structure alignment algorithm as the main engine for detecting structure similarities between the protein of interest and library spheres. An all-atom structure similarity metric ensures that sidechain placement is taken into account in the PDBspheres primary assessment of confidence in structural matches. In this paper, we (1) describe the PDBspheres method, (2) demonstrate how PDBspheres can be used to detect and characterize binding sites in protein structures, (3) compare PDBspheres use for binding site prediction with seven other binding site prediction methods using a curated dataset of 2,528 ligand-bound and ligand-free crystal structures, and (4) use PDBspheres to cluster pockets and assess structural similarities among protein binding sites of the 4,876 structures in the refined set of PDBbind 2019 dataset. The PDBspheres library is made publicly available for download at https://proteinmodel.org/AS2TS/PDBspheres

3D U-Net: A voxel-based method in binding site prediction of protein structure

Binding site prediction for new proteins is important in structure-based drug design. The identified binding sites may be helpful in the development of treatments for new viral outbreaks in the world when there is no information available about their pockets with COVID-19 being a case in point. Identification of the pockets using computational methods, as an alternative method, has recently attracted much interest. In this study, the binding site prediction is viewed as a semantic segmentation problem. An improved 3D version of the U-Net model based on the dice loss function is utilized to predict the binding sites accurately. The performance of the proposed model on the independent test datasets and SARS-COV-2 shows the segmentation model could predict the binding sites with a more accurate shape than the recently published deep learning model, i.e. DeepSite. Therefore, the model may help predict the binding sites of proteins and could be used in drug design for novel proteins.

Predicting binding sites from unbound versus bound protein structures

We present the application of seven binding-site prediction algorithms to a meticulously curated dataset of ligand-bound and ligand-free crystal structures for 304 unique protein sequences (2528 crystal structures). We probe the influence of starting protein structures on the results of binding-site prediction, so the dataset contains a minimum of two ligand-bound and two ligand-free structures for each protein. We use this dataset in a brief survey of five geometry-based, one energy-based, and one machine-learning-based methods: Surfnet, Ghecom, LIGSITEcsc, Fpocket, Depth, AutoSite, and Kalasanty. Distributions of the F scores and Matthew’s correlation coefficients for ligand-bound versus ligand-free structure performance show no statistically significant difference in structure type versus performance for most methods. Only Fpocket showed a statistically significant but low magnitude enhancement in performance for holo structures. Lastly, we found that most methods will succeed on some crystal structures and fail on others within the same protein family, despite all structures being relatively high-quality structures with low structural variation. We expected better consistency across varying protein conformations of the same sequence. Interestingly, the success or failure of a given structure cannot be predicted by quality metrics such as resolution, Cruickshank Diffraction Precision index, or unresolved residues. Cryptic sites were also examined.

aPRBind: protein–RNA interface prediction by combining sequence and I-TASSER model-based structural features learned with convolutional neural networks

Motivation Protein–RNA interactions play a critical role in various biological processes. The accurate prediction of RNA-binding residues in proteins has been one of the most challenging and intriguing problems in the field of computational biology. The existing methods still have a relatively low accuracy especially for the sequence-based ab-initio methods. Results In this work, we propose an approach aPRBind, a convolutional neural network-based ab-initio method for RNA-binding residue prediction. aPRBind is trained with sequence features and structural ones (particularly including residue dynamics information and residue–nucleotide propensity developed by us) that are extracted from the predicted structures by I-TASSER. The analysis of feature contributions indicates the sequence features are most important, followed by dynamics information, and the sequence and structural features are complementary in binding site prediction. The performance comparison of our method with other peer ones on benchmark dataset shows that aPRBind outperforms some state-of-the-art ab-initio methods. Additionally, aPRBind can give a better prediction for the modeled structures with TM-score≥0.5, and meanwhile since the structural features are not very sensitive to the refined 3D structures, aPRBind has only a marginal dependence on the accuracy of the structure model, which allows aPRBind to be applied to the RNA-binding site prediction for the modeled or unbound structures. Availability and implementation The source code is available at https://github.com/ChunhuaLiLab/aPRbind. Supplementary information Supplementary data are available at Bioinformatics online.