RESUMEN

BACKGROUND: Locating small molecule binding sites in target proteins, in the resolution of either pocket or residue, is critical in many drug-discovery scenarios. Since it is not always easy to find such binding sites using conventional methods, different deep learning methods to predict binding sites out of protein structures have been developed in recent years. The existing deep learning based methods have several limitations, including (1) the inefficiency of the CNN-only architecture, (2) loss of information due to excessive post-processing, and (3) the under-utilization of available data sources. METHODS: We present a new model architecture and training method that resolves the aforementioned problems. First, by layering geometric self-attention units on top of residue-level 3D CNN outputs, our model overcomes the problems of CNN-only architectures. Second, by configuring the fundamental units of computation as residues and pockets instead of voxels, our method reduced the information loss from post-processing. Lastly, by employing inter-resolution transfer learning and homology-based augmentation, our method maximizes the utilization of available data sources to a significant extent. RESULTS: The proposed method significantly outperformed all state-of-the-art baselines regarding both resolutions-pocket and residue. An ablation study demonstrated the indispensability of our proposed architecture, as well as transfer learning and homology-based augmentation, for achieving optimal performance. We further scrutinized our model's performance through a case study involving human serum albumin, which demonstrated our model's superior capability in identifying multiple binding sites of the protein, outperforming the existing methods. CONCLUSIONS: We believe that our contribution to the literature is twofold. Firstly, we introduce a novel computational method for binding site prediction with practical applications, substantiated by its strong performance across diverse benchmarks and case studies. Secondly, the innovative aspects in our method- specifically, the design of the model architecture, inter-resolution transfer learning, and homology-based augmentation-would serve as useful components for future work.

Asunto(s)

RESUMEN

Revealing protein binding sites with other molecules, such as nucleic acids, peptides, or small ligands, sheds light on disease mechanism elucidation and novel drug design. With the explosive growth of proteins in sequence databases, how to accurately and efficiently identify these binding sites from sequences becomes essential. However, current methods mostly rely on expensive multiple sequence alignments or experimental protein structures, limiting their genome-scale applications. Besides, these methods haven't fully explored the geometry of the protein structures. Here, we propose GPSite, a multi-task network for simultaneously predicting binding residues of DNA, RNA, peptide, protein, ATP, HEM, and metal ions on proteins. GPSite was trained on informative sequence embeddings and predicted structures from protein language models, while comprehensively extracting residual and relational geometric contexts in an end-to-end manner. Experiments demonstrate that GPSite substantially surpasses state-of-the-art sequence-based and structure-based approaches on various benchmark datasets, even when the structures are not well-predicted. The low computational cost of GPSite enables rapid genome-scale binding residue annotations for over 568,000 sequences, providing opportunities to unveil unexplored associations of binding sites with molecular functions, biological processes, and genetic variants. The GPSite webserver and annotation database can be freely accessed at https://bio-web1.nscc-gz.cn/app/GPSite.

Asunto(s)

RESUMEN

Several diverse proteins possess similar binding sites. Protein binding site comparison provides valuable insights for the drug discovery and development. Binding site similarities are useful in understanding polypharmacology, identifying potential off-targets and repurposing of known drugs. Many binding site analysis and comparison methods are available today, however, these methods may not be adequate to explain variation in the activity of a drug or a small molecule against a number of similar proteins. Water molecules surrounding the protein surface contribute to structure and function of proteins. Water molecules form diverse types of hydrogen-bonded cyclic water-ring networks known as topological water networks (TWNs). Analysis of TWNs in binding site of proteins may improve understanding of the characteristics of binding sites. We propose TWN-based residue encoding (TWN-RENCOD), a novel binding site comparison method which compares the aqueous environment in binding sites of similar proteins. As compared to other existing methods, results obtained using our method correlated better with differences in wide range of activity of a known drug (Sunitinib) against nine different protein kinases (KIT, PDGFRA, VEGFR2, PHKG2, ITK, HPK1, MST3, PAK6 and CDK2).

RESUMEN

Though AlphaFold2 has attained considerably high precision on protein structure prediction, it is reported that directly inputting coordinates into deep learning networks cannot achieve desirable results on downstream tasks. Thus, how to process and encode the predicted results into effective forms that deep learning models can understand to improve the performance of downstream tasks is worth exploring. In this study, we tested the effects of five processing strategies of coordinates on two single-sequence protein binding site prediction tasks. These five strategies are spatial filtering, the singular value decomposition of a distance map, calculating the secondary structure feature, and the relative accessible surface area feature of proteins. The computational experiment results showed that all strategies were suitable and effective methods to encode structural information for deep learning models. In addition, by performing a case study of a mutated protein, we showed that the spatial filtering strategy could introduce structural changes into HHblits profiles and deep learning networks when protein mutation happens. In sum, this work provides new insight into the downstream tasks of protein-molecule interaction prediction, such as predicting the binding residues of proteins and estimating the effects of mutations.

RESUMEN

Protein mapping distributes many copies of different molecular probes on the surface of a target protein in order to determine binding hot spots, regions that are highly preferable for ligand binding. While mapping of X-ray structures by the FTMap server is inherently static, this limitation can be overcome by the simultaneous analysis of multiple structures of the protein. FTMove is an automated web server that implements this approach. From the input of a target protein, by PDB code, the server identifies all structures of the protein available in the PDB, runs mapping on them, and combines the results to form binding hot spots and binding sites. The user may also upload their own protein structures, bypassing the PDB search for similar structures. Output of the server consists of the consensus binding sites and the individual mapping results for each structure - including the number of probes located in each binding site, for each structure. This level of detail allows the users to investigate how the strength of a binding site relates to the protein conformation, other binding sites, and the presence of ligands or mutations. In addition, the structures are clustered on the basis of their binding properties. The use of FTMove is demonstrated by application to 22 proteins with known allosteric binding sites; the orthosteric and allosteric binding sites were identified in all but one case, and the sites were typically ranked among the top five. The FTMove server is publicly available at https://ftmove.bu.edu.

Asunto(s)

RESUMEN

Proteins are capable of highly specific interactions and are responsible for a wide range of functions, making them attractive in the pursuit of new therapeutic options. Previous studies focusing on overall geometry of protein-protein interfaces, however, concluded that PPI interfaces were generally flat. More recently, this idea has been challenged by their structural and thermodynamic characterisation, suggesting the existence of concave binding sites that are closer in character to traditional small-molecule binding sites, rather than exhibiting complete flatness. Here, we present a large-scale analysis of binding geometry and physicochemical properties of all protein-protein interfaces available in the Protein Data Bank. In this review, we provide a comprehensive overview of the protein-protein interface landscape, including evidence that even for overall larger, more flat interfaces that utilize discontinuous interacting regions, small and potentially druggable pockets are utilized at binding sites.

Asunto(s)

RESUMEN

Identifying binding sites on the protein surface is an important part of computer-assisted drug design processes. Reliable prediction of binding sites not only assists with docking algorithms, but it can also explain the possible side-effects of a potential drug as well as its efficiency. In this work, we propose a novel workflow for predicting possible binding sites of a ligand on a protein surface. We use proteins from the PDBbind and sc-PDB databases, from which we combine available ligand information for similar proteins using all the possible ligands rather than only a special sub-selection to generalize the work of existing research. After performing protein clustering and merging of ligands of similar proteins, we use a three-dimensional convolutional neural network that takes into account the spatial structure of a protein. Lastly, we combine ligandability predictions for points on protein surfaces into joint binding sites. Analysis of our model's performance shows that its achieved sensitivity is 0.829, specificity is 0.98, and F1 score is 0.517, and that for 54% of larger and pharmacologically relevant binding sites, the distance between their real and predicted centers amounts to less than 4 Å.

RESUMEN

An important question is how well the models submitted to CASP retain the properties of target structures. We investigate several properties related to binding. First we explore the binding of small molecules as probes, and count the number of interactions between each residue and such probes, resulting in a binding fingerprint. The similarity between two fingerprints, one for the X-ray structure and the other for a model, is determined by calculating their correlation coefficient. The fingerprint similarity weakly correlates with global measures of accuracy, and GDT_TS higher than 80 is a necessary but not sufficient condition for the conservation of surface binding properties. The advantage of this approach is that it can be carried out without information on potential ligands and their binding sites. The latter information was available for a few targets, and we explored whether the CASP14 models can be used to predict binding sites and to dock small ligands. Finally, we tested the ability of models to reproduce protein-protein interactions by docking both the X-ray structures and the models to their interaction partners in complexes. The analysis showed that in CASP14 the quality of individual domain models is approaching that offered by X-ray crystallography, and hence such models can be successfully used for the identification of binding and regulatory sites, as well as for assembling obligatory protein-protein complexes. Success of ligand docking, however, often depends on fine details of the binding interface, and thus may require accounting for conformational changes by simulation methods.

Asunto(s)

RESUMEN

The expansion of three-dimensional protein structures and enhanced computing power have significantly facilitated our understanding of protein sequence/structure/function relationships. A challenge in structural genomics is to predict the function of uncharacterized proteins. Protein function deconvolution based on global sequence or structural homology is impracticable when a protein relates to no other proteins with known function, and in such cases, functional relationships can be established by detecting their local ligand binding site similarity. Here, we introduce a sequence order-independent comparison algorithm, PocketShape, for structural proteome-wide exploration of protein functional site by fully considering the geometry of the backbones, orientation of the sidechains, and physiochemical properties of the pocket-lining residues. PocketShape is efficient in distinguishing similar from dissimilar ligand binding site pairs by retrieving 99.3% of the similar pairs while rejecting 100% of the dissimilar pairs on a dataset containing 1538 binding site pairs. This method successfully classifies 83 enzyme structures with diverse functions into 12 clusters, which is highly in accordance with the actual structural classification of proteins classification. PocketShape also achieves superior performances than other methods in protein profiling based on experimental data. Potential new applications for representative SARS-CoV-2 drugs Remdesivir and 11a are predicted. The high accuracy and time-efficient characteristics of PocketShape will undoubtedly make it a promising complementary tool for proteome-wide protein function inference and drug repurposing study.

Asunto(s)

RESUMEN

We study the models submitted to round 12 of the Critical Assessment of protein Structure Prediction (CASP) experiment to assess how well the binding properties are conserved when the X-ray structures of the target proteins are replaced by their models. To explore small molecule binding we generate distributions of molecular probes - which are fragment-sized organic molecules of varying size, shape, and polarity - around the protein, and count the number of interactions between each residue and the probes, resulting in a vector of interactions we call a binding fingerprint. The similarity between two fingerprints, one for the X-ray structure and the other for a model of the protein, is determined by calculating the correlation coefficient between the two vectors. The resulting correlation coefficients are shown to correlate with global measures of accuracy established in CASP, and the relationship yields an accuracy threshold that has to be reached for meaningful binding surface conservation. The clusters formed by the probe molecules reliably predict binding hot spots and ligand binding sites in both X-ray structures and reasonably accurate models of the target, but ensembles of models may be needed for assessing the availability of proper binding pockets. We explored ligand docking to the few targets that had bound ligands in the X-ray structure. More targets were available to assess the ability of the models to reproduce protein-protein interactions by docking both the X-ray structures and models to their interaction partners in complexes. It was shown that this application is more difficult than finding small ligand binding sites, and the success rates heavily depend on the local structure in the potential interface. In particular, predicted conformations of flexible loops are frequently incorrect in otherwise highly accurate models, and may prevent predicting correct protein-protein interactions.

RESUMEN

Infrared (IR) spectroscopy is a well-established technique for probing the structure, behavior, and surroundings of molecules in their native environments. Its characteristics-most specifically high structural sensitivity, ready applicability to aqueous samples, and broad availability-make it a valuable enzymological technique, particularly for the interrogation of ion binding sites. While IR spectroscopy of the "garden variety" (steady state at room temperature with wild-type proteins) is versatile and powerful in its own right, the combination of IR spectroscopy with specialized experimental schemes for leveraging ultrafast time resolution, protein labeling, and other enhancements further extends this utility. This book chapter provides the fundamental physical background and literature context essential for harnessing IR spectroscopy in the general context of enzymology with specific focus on interrogation of ion binding. Studies of lanthanide ions binding to calmodulin are highlighted as illustrative examples of this process. Appropriate sample preparation, data collection, and spectral interpretation are discussed from a detail-oriented and practical perspective with the goal of facilitating the reader's rapid progression from reading words in a book to collecting and analyzing their own data in the lab.

Asunto(s)

RESUMEN

DNA/RNA motif mining is the foundation of gene function research. The DNA/RNA motif mining plays an extremely important role in identifying the DNA- or RNA-protein binding site, which helps to understand the mechanism of gene regulation and management. For the past few decades, researchers have been working on designing new efficient and accurate algorithms for mining motif. These algorithms can be roughly divided into two categories: the enumeration approach and the probabilistic method. In recent years, machine learning methods had made great progress, especially the algorithm represented by deep learning had achieved good performance. Existing deep learning methods in motif mining can be roughly divided into three types of models: convolutional neural network (CNN) based models, recurrent neural network (RNN) based models, and hybrid CNN-RNN based models. We introduce the application of deep learning in the field of motif mining in terms of data preprocessing, features of existing deep learning architectures and comparing the differences between the basic deep learning models. Through the analysis and comparison of existing deep learning methods, we found that the more complex models tend to perform better than simple ones when data are sufficient, and the current methods are relatively simple compared with other fields such as computer vision, language processing (NLP), computer games, etc. Therefore, it is necessary to conduct a summary in motif mining by deep learning, which can help researchers understand this field.

Asunto(s)

RESUMEN

Mutual information and entropy transfer analysis employed on two inactive states of human beta-2 adrenergic receptor (ß2 -AR) unraveled distinct communication pathways. Previously, a so-called "highly" inactive state of the receptor was observed during 1.5 microsecond long molecular dynamics simulation where the largest intracellular loop (ICL3) was swiftly packed onto the G-protein binding cavity, becoming entirely inaccessible. Mutual information quantifying the degree of correspondence between backbone-Cα fluctuations was mostly shared between intra- and extra-cellular loop regions in the original inactive state, but shifted to entirely different regions in this latest inactive state. Interestingly, the largest amount of mutual information was always shared among the mobile regions. Irrespective of the conformational state, polar residues always contributed more to mutual information than hydrophobic residues, and also the number of polar-polar residue pairs shared the highest degree of mutual information compared to those incorporating hydrophobic residues. Entropy transfer, quantifying the correspondence between backbone-Cα fluctuations at different timesteps, revealed a distinctive pathway directed from the extracellular site toward intracellular portions in this recently exposed inactive state for which the direction of information flow was the reverse of that observed in the original inactive state where the mobile ICL3 and its intracellular surroundings drove the future fluctuations of extracellular regions.

Asunto(s)

RESUMEN

Prediction of binding affinity of proteins and small molecules is a key step in drug design, and the location of binding sites is crucial for affinity prediction and molecular docking. In order to improve the accuracy of binding site prediction, a method called FRSite which improves the Faster R-CNN for protein binding site prediction is proposed in this paper. Multi-channel descriptors for proteins are generated to three dimensional (3D) girds and fed into the proposed Region Proposal Network (RPN-3D) network for potential proposals detection. Moreover, a 3D classifier is used to predict the bounding box of the binding site for a protein, and could also predict the center and size of the site. It can be seen from our comparative experiments that the proposed method can assist drug design.

Asunto(s)

RESUMEN

The retinoic acid receptor-related orphan receptor (ROR) γt receptor is a member of nuclear receptors, which is indispensable for the expression of pro-inflammatory cytokine IL-17. RORγt has been established as a drug target to design and discover novel treatments for multiple inflammatory and immunological diseases. It is important to elucidate the molecular mechanisms of how RORγt is activated by an agonist, and how the transcription function of RORγt is interrupted by an inverse agonist. In this study we performed molecular dynamics simulations on four different RORγt systems, i.e., the apo protein, protein bound with agonist, protein bound with inverse agonist in the orthosteric-binding pocket, and protein bound with inverse agonist in the allosteric-binding pocket. We found that the orthosteric-binding pocket in the apo-form RORγt was mostly open, confirming that apo-form RORγt was constitutively active and could be readily activated (ca. tens of nanoseconds scale). The tracked data from MD simulations supported that RORγt could be activated by an agonist binding at the orthosteric-binding pocket, because the bound agonist helped to enhance the triplet His479-Tyr502-Phe506 interactions and stabilized H12 structure. The stabilized H12 helped RORγt to form the protein-binding site, and therefore made the receptor ready to recruit a coactivator molecule. We also showed that transcription function of RORγt could be interrupted by the binding of inverse agonist at the orthosteric-binding pocket or at the allosteric-binding site. After the inverse agonist was bound, H12 either structurally collapsed, or reorientated to a different position, at which the presumed protein-binding site was not able to be formed.

Asunto(s)

RESUMEN

Reporter genes have contributed to advancements in molecular biology. Binding of an upstream regulatory protein to a downstream reporter promoter allows quantification of the activity of the upstream protein produced from the corresponding gene. In studies of synthetic biology, analyses of reporter gene activities ensure control of the cell with synthetic genetic circuits, as achieved using a combination of in silico and in vivo experiments. However, unexpected effects of downstream reporter genes on upstream regulatory genes may interfere with in vivo observations. This phenomenon is termed as retroactivity. Using in silico and in vivo experiments, we found that a different copy number of regulatory protein-binding sites in a downstream gene altered the upstream dynamics, suggesting retroactivity of reporters in this synthetic genetic oscillator. Furthermore, by separating the two sources of retroactivity (titration of the component and competition for degradation), we showed that, in the dual-feedback oscillator, the level of the fluorescent protein reporter competing for degradation with the circuits' components is important for the stability of the oscillations. Altogether, our results indicate that the selection of reporter promoters using a combination of in silico and in vivo experiments is essential for the advanced design of genetic circuits.

RESUMEN

De novo motif discovery in biological sequences is an important and computationally challenging problem. A myriad of algorithms have been developed to solve this problem with varying success, but it can be difficult for even a small number of these tools to reach a consensus. Because individual tools can be better suited for specific scenarios, an ensemble tool that combines the results of many algorithms can yield a more confident and complete result. We present a novel and fast tool ensemble MCAT (Motif Combining and Association Tool) for de novo motif discovery by combining six state-of-the-art motif discovery tools (MEME, BioProspector, DECOD, XXmotif, Weeder, and CMF). We apply MCAT to data sets with DNA sequences that come from various species and compare our results with two well-established ensemble motif-finding tools, EMD and DynaMIT. The experimental results show that MCAT is able to identify exact match motifs in DNA sequences efficiently, and it has a significantly better performance in practice.

Asunto(s)

RESUMEN

PURPOSE: Von Hippel-Lindau (VHL) disease is a rare hereditary cancer syndrome that reduces life expectancy. We aimed to construct a more valuable genotype-phenotype correlation based on alterations in VHL protein (pVHL). METHODS: VHL patients (n = 339) were recruited and grouped based on mutation types: HIF-α binding site missense (HM) mutations, non-HIF-α binding site missense (nHM) mutations, and truncating (TR) mutations. Age-related risks of VHL-associated tumors and patient survival were compared. RESULTS: Missense mutations conferred an increased risk of pheochromocytoma (HR = 1.854, p = 0.047) compared with truncating mutations. The risk of pheochromocytoma was lower in the HM group than in the nHM group (HR = 0.298, p = 0.003) but was similar between HM and TR groups (HR = 0.901, p = 0.810). Patients in the nHM group had a higher risk of pheochromocytoma (HR = 3.447, p < 0.001) and lower risks of central nervous system hemangioblastoma (CHB) (HR = 0.700, p = 0.045), renal cell carcinoma (HR = 0.610, p = 0.024), and pancreatic tumor (HR = 0.382, p < 0.001) than those in the combined HM and TR (HMTR) group. Moreover, nHM mutations were independently associated with better overall survival (HR = 0.345, p = 0.005) and CHB-specific survival (HR = 0.129, p = 0.005) than HMTR mutations. CONCLUSION: The modified genotype-phenotype correlation links VHL gene mutation, substrate binding site, and phenotypic diversity (penetrance and survival), and provides more accurate information for genetic counseling and pathogenesis studies.

Asunto(s)

RESUMEN

In this article, we present a new approach to expand the range of application of protein-ligand docking methods in the prediction of the interaction of coordination complexes (i.e., metallodrugs, natural and artificial cofactors, etc.) with proteins. To do so, we assume that, from a pure computational point of view, hydrogen bond functions could be an adequate model for the coordination bonds as both share directionality and polarity aspects. In this model, docking of metalloligands can be performed without using any geometrical constraints or energy restraints. The hard work consists in generating the convenient atom types and scoring functions. To test this approach, we applied our model to 39 high-quality X-ray structures with transition and main group metal complexes bound via a unique coordination bond to a protein. This concept was implemented in the protein-ligand docking program GOLD. The results are in very good agreement with the experimental structures: the percentage for which the RMSD of the simulated pose is smaller than the X-ray spectra resolution is 92.3% and the mean value of RMSD is < 1.0 Å. Such results also show the viability of the method to predict metal complexes-proteins interactions when the X-ray structure is not available. This work could be the first step for novel applicability of docking techniques in medicinal and bioinorganic chemistry and appears generalizable enough to be implemented in most protein-ligand docking programs nowadays available. © 2017 Wiley Periodicals, Inc.

Asunto(s)

RESUMEN

Due to Ca2+ -dependent binding and the sequence diversity of Calmodulin (CaM) binding proteins, identifying CaM interactions and binding sites in the wet-lab is tedious and costly. Therefore, computational methods for this purpose are crucial to the design of such wet-lab experiments. We present an algorithm suite called CaMELS (CalModulin intEraction Learning System) for predicting proteins that interact with CaM as well as their binding sites using sequence information alone. CaMELS offers state of the art accuracy for both CaM interaction and binding site prediction and can aid biologists in studying CaM binding proteins. For CaM interaction prediction, CaMELS uses protein sequence features coupled with a large-margin classifier. CaMELS models the binding site prediction problem using multiple instance machine learning with a custom optimization algorithm which allows more effective learning over imprecisely annotated CaM-binding sites during training. CaMELS has been extensively benchmarked using a variety of data sets, mutagenic studies, proteome-wide Gene Ontology enrichment analyses and protein structures. Our experiments indicate that CaMELS outperforms simple motif-based search and other existing methods for interaction and binding site prediction. We have also found that the whole sequence of a protein, rather than just its binding site, is important for predicting its interaction with CaM. Using the machine learning model in CaMELS, we have identified important features of protein sequences for CaM interaction prediction as well as characteristic amino acid sub-sequences and their relative position for identifying CaM binding sites. Python code for training and evaluating CaMELS together with a webserver implementation is available at the URL: http://faculty.pieas.edu.pk/fayyaz/software.html#camels.

Asunto(s)