Binding affinity prediction with property-encoded shape distribution signatures.
Ontology highlight
ABSTRACT:
SUBMITTER: Das S
PROVIDER: S-EPMC2846646 | biostudies-literature | 2010 Feb
REPOSITORIES: biostudies-literature
Publications
Binding affinity prediction with property-encoded shape distribution signatures.
Das Sourav S Krein Michael P MP Breneman Curt M CM
Journal of chemical information and modeling 20100201 2
We report the use of the molecular signatures known as "property-encoded shape distributions" (PESD) together with standard support vector machine (SVM) techniques to produce validated models that can predict the binding affinity of a large number of protein ligand complexes. This "PESD-SVM" method uses PESD signatures that encode molecular shapes and property distributions on protein and ligand surfaces as features to build SVM models that require no subjective feature selection. A simple proto ...[more]
Similar Datasets
Project description:Patterns in shape and property distributions on the surface of binding sites are often conserved across functional proteins without significant conservation of the underlying amino-acid residues. To explore similarities of these sites from the viewpoint of a ligand, a sequence and fold-independent method was created to rapidly and accurately compare binding sites of proteins represented by property-mapped triangulated Gauss-Connolly surfaces. Within this paradigm, signatures for each binding site surface are produced by calculating their property-encoded shape distributions (PESD), a measure of the probability that a particular property will be at a specific distance to another on the molecular surface. Similarity between the signatures can then be treated as a measure of similarity between binding sites. As postulated, the PESD method rapidly detected high levels of similarity in binding site surface characteristics even in cases where there was very low similarity at the sequence level. In a screening experiment involving each member of the PDBBind 2005 data set as a query against the rest of the set, PESD was able to retrieve a binding site with identical E.C. (Enzyme Commission) numbers as the top match in 79.5% of cases. The ability of the method in detecting similarity in binding sites with low sequence conservations was compared to state-of-the-art binding site comparison methods.
| S-EPMC2804948 | biostudies-literature
Project description:Computational methods for predicting ligand affinity where no protein structure is known generally take the form of regression analysis based on molecular features that have only a tangential relationship to a protein/ligand binding event. Such methods have limited utility when structural variation moves beyond congeneric series. We present a novel approach based on the multiple-instance learning method of Compass, where a physical model of a binding site is induced from ligands and their corresponding activity data. The model consists of molecular fragments that can account for multiple positions of literal protein residues. We demonstrate the method on 5HT1a ligands by training on a series with limited scaffold variation and testing on numerous ligands with variant scaffolds. Predictive error was between 0.5 and 1.0 log units (0.7-1.4 kcal/mol), with statistically significant rank correlations. Accurate activity predictions of novel ligands were demonstrated using a validation approach where a small number of ligands of limited structural variation known at a fixed time point were used to make predictions on a blind test set of widely varying molecules, some discovered at a much later time point.
| S-EPMC2766559 | biostudies-literature
Project description:MotivationThe identification of novel drug-target (DT) interactions is a substantial part of the drug discovery process. Most of the computational methods that have been proposed to predict DT interactions have focused on binary classification, where the goal is to determine whether a DT pair interacts or not. However, protein-ligand interactions assume a continuum of binding strength values, also called binding affinity and predicting this value still remains a challenge. The increase in the affinity data available in DT knowledge-bases allows the use of advanced learning techniques such as deep learning architectures in the prediction of binding affinities. In this study, we propose a deep-learning based model that uses only sequence information of both targets and drugs to predict DT interaction binding affinities. The few studies that focus on DT binding affinity prediction use either 3D structures of protein-ligand complexes or 2D features of compounds. One novel approach used in this work is the modeling of protein sequences and compound 1D representations with convolutional neural networks (CNNs).ResultsThe results show that the proposed deep learning based model that uses the 1D representations of targets and drugs is an effective approach for drug target binding affinity prediction. The model in which high-level representations of a drug and a target are constructed via CNNs achieved the best Concordance Index (CI) performance in one of our larger benchmark datasets, outperforming the KronRLS algorithm and SimBoost, a state-of-the-art method for DT binding affinity prediction.Availability and implementationhttps://github.com/hkmztrk/DeepDTA.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC6129291 | biostudies-literature
Project description:The optimization of compounds' binding affinity for a biological target is a crucial aspect of the drug development process. Being able to accurately predict binding energies in advance of synthesizing compounds would have a massive impact on the speed of the drug discovery process. The ideal binding affinity prediction method should combine accuracy, reliability, and speed. In this paper, we present SophosQM, a quantum mechanics (QM)-based approach, which can accurately predict the binding affinities of compounds to proteins. The binding affinity predictive models generated by SophosQM are based on the fragment molecular orbital (FMO) method to estimate the enthalpic component of the binding free energy, and a macroscopic descriptor, clog P, is used as an approximation of the entropic component. The affinity prediction is performed using multilinear regression, fitting the experimental values against the FMO-computed enthalpic term and clog P. The quality of the prediction can be assessed in terms of the correlation coefficient between experimental and predicted values. In this work, the method's reliability and accuracy are exemplified by applying SophosQM to 70 compounds binding to six different targets of pharmaceutical relevance. Overall, the results show a very satisfactory performance with a global correlation coefficient in the order of 0.9. Our predictions also show a satisfactory performance compared to data based on free energy perturbation. Finally, SophosQM can also be applied in high-throughput mode by using semiempirical QM methods to evaluate large portions of chemical space, while retaining a good level of accuracy, but decreasing the computing time to just a few seconds per compound.
| S-EPMC10157843 | biostudies-literature
Project description:Interactions between RNA and proteins are pervasive in biology, driving fundamental processes such as protein translation and participating in the regulation of gene expression. Modeling the energies of RNA-protein interactions is therefore critical for understanding and repurposing living systems but has been hindered by complexities unique to RNA-protein binding. Here, we bring together several advances to complete a calculation framework for RNA-protein binding affinities, including a unified free energy function for bound complexes, automated Rosetta modeling of mutations, and use of secondary structure-based energetic calculations to model unbound RNA states. The resulting Rosetta-Vienna RNP-ΔΔG method achieves root-mean-squared errors (RMSEs) of 1.3 kcal/mol on high-throughput MS2 coat protein-RNA measurements and 1.5 kcal/mol on an independent test set involving the signal recognition particle, human U1A, PUM1, and FOX-1. As a stringent test, the method achieves RMSE accuracy of 1.4 kcal/mol in blind predictions of hundreds of human PUM2-RNA relative binding affinities. Overall, these RMSE accuracies are significantly better than those attained by prior structure-based approaches applied to the same systems. Importantly, Rosetta-Vienna RNP-ΔΔG establishes a framework for further improvements in modeling RNA-protein binding that can be tested by prospective high-throughput measurements on new systems.
| S-EPMC6486753 | biostudies-literature
Project description:The prediction of the binding free energy between a ligand and a protein is an important component in the virtual screening and lead optimization of ligands for drug discovery. To determine the quality of current binding free energy estimation programs, we examined FlexX, X-Score, AutoDock, and BLEEP for their performance in binding free energy prediction in various situations including cocrystallized complex structures, cross docking of ligands to their non-cocrystallized receptors, docking of thermally unfolded receptor decoys to their ligands, and complex structures with "randomized" ligand decoys. In no case was there a satisfactory correlation between the experimental and estimated binding free energies over all the datasets tested. Meanwhile, a strong correlation between ligand molecular weight-binding affinity correlation and experimental predicted binding affinity correlation was found. Sometimes the programs also correctly ranked ligands' binding affinities even though native interactions between the ligands and their receptors were essentially lost because of receptor deformation or ligand randomization, and the programs could not decisively discriminate randomized ligand decoys from their native ligands; this suggested that the tested programs miss important components for the accurate capture of specific ligand binding interactions.
| S-EPMC2702145 | biostudies-literature
Project description:Protein function prediction is an active area of research in computational biology. Function prediction can help biologists make hypotheses for characterization of genes and help interpret biological assays, and thus is a productive area for collaboration between experimental and computational biologists. Among various function prediction methods, predicting binding ligand molecules for a target protein is an important class because ligand binding events for a protein are usually closely intertwined with the proteins' biological function, and also because predicted binding ligands can often be directly tested by biochemical assays. Binding ligand prediction methods can be classified into two types: those which are based on protein-protein (or pocket-pocket) comparison, and those that compare a target pocket directly to ligands. Recently, our group proposed two computational binding ligand prediction methods, Patch-Surfer, which is a pocket-pocket comparison method, and PL-PatchSurfer, which compares a pocket to ligand molecules. The two programs apply surface patch-based descriptions to calculate similarity or complementarity between molecules. A surface patch is characterized by physicochemical properties such as shape, hydrophobicity, and electrostatic potentials. These properties on the surface are represented using three-dimensional Zernike descriptors (3DZD), which are based on a series expansion of a 3 dimensional function. Utilizing 3DZD for describing the physicochemical properties has two main advantages: (1) rotational invariance and (2) fast comparison. Here, we introduce Patch-Surfer and PL-PatchSurfer with an emphasis on PL-PatchSurfer, which is more recently developed. Illustrative examples of PL-PatchSurfer performance on binding ligand prediction as well as virtual drug screening are also provided.
| S-EPMC4718779 | biostudies-literature
Project description:BACKGROUND: Various pattern-based methods exist that use in vitro or in silico affinity profiles for classification and functional examination of proteins. Nevertheless, the connection between the protein affinity profiles and the structural characteristics of the binding sites is still unclear. Our aim was to investigate the association between virtual drug screening results (calculated binding free energy values) and the geometry of protein binding sites. Molecular Affinity Fingerprints (MAFs) were determined for 154 proteins based on their molecular docking energy results for 1,255 FDA-approved drugs. Protein binding site geometries were characterized by 420 PocketPicker descriptors. The basic underlying component structure of MAFs and binding site geometries, respectively, were examined by principal component analysis; association between principal components extracted from these two sets of variables was then investigated by canonical correlation and redundancy analyses. RESULTS: PCA analysis of the MAF variables provided 30 factors which explained 71.4% of the total variance of the energy values while 13 factors were obtained from the PocketPicker descriptors which cumulatively explained 94.1% of the total variance. Canonical correlation analysis resulted in 3 statistically significant canonical factor pairs with correlation values of 0.87, 0.84 and 0.77, respectively. Redundancy analysis indicated that PocketPicker descriptor factors explain 6.9% of the variance of the MAF factor set while MAF factors explain 15.9% of the total variance of PocketPicker descriptor factors. Based on the salient structures of the factor pairs, we identified a clear-cut association between the shape and bulkiness of the drug molecules and the protein binding site descriptors. CONCLUSIONS: This is the first study to investigate complex multivariate associations between affinity profiles and the geometric properties of protein binding sites. We found that, except for few specific cases, the shapes of the binding pockets have relatively low weights in the determination of the affinity profiles of proteins. Since the MAF profile is closely related to the target specificity of ligand binding sites we can conclude that the shape of the binding site is not a pivotal factor in selecting drug targets. Nonetheless, based on strong specific associations between certain MAF profiles and specific geometric descriptors we identified, the shapes of the binding sites do have a crucial role in virtual drug design for certain drug categories, including morphine derivatives, benzodiazepines, barbiturates and antihistamines.
| S-EPMC2972294 | biostudies-literature
Project description:The combination of physicochemical properties and energetic parameters derived from protein-ligand complexes play a vital role in determining the biological activity of a molecule. In the present work, protein-ligand interaction energy along with logP values was used to predict the experimental log (IC50) values of 25 different kinase-inhibitors using multiple regressions which gave a correlation coefficient of 0.93. The regression equation obtained was tested on 93 kinase-inhibitor complexes and an average deviation of 0.92 from the experimental log IC50 values was shown. The same set of descriptors was used to predict binding affinities for a test set of five individual kinase families, with correlation values > 0.9. We show that the protein-ligand interaction energies and partition coefficient values form the major deterministic factors for binding affinity of the ligand for its receptor.
| S-EPMC5267961 | biostudies-literature
Project description:Background Several computational advances have been achieved in the drug discovery field, promoting the identification of novel drug–target interactions and new leads. However, most of these methodologies have been overlooking the importance of providing explanations to the decision-making process of deep learning architectures. In this research study, we explore the reliability of convolutional neural networks (CNNs) at identifying relevant regions for binding, specifically binding sites and motifs, and the significance of the deep representations extracted by providing explanations to the model’s decisions based on the identification of the input regions that contributed the most to the prediction. We make use of an end-to-end deep learning architecture to predict binding affinity, where CNNs are exploited in their capacity to automatically identify and extract discriminating deep representations from 1D sequential and structural data. Results The results demonstrate the effectiveness of the deep representations extracted from CNNs in the prediction of drug–target interactions. CNNs were found to identify and extract features from regions relevant for the interaction, where the weight associated with these spots was in the range of those with the highest positive influence given by the CNNs in the prediction. The end-to-end deep learning model achieved the highest performance both in the prediction of the binding affinity and on the ability to correctly distinguish the interaction strength rank order when compared to baseline approaches. Conclusions This research study validates the potential applicability of an end-to-end deep learning architecture in the context of drug discovery beyond the confined space of proteins and ligands with determined 3D structure. Furthermore, it shows the reliability of the deep representations extracted from the CNNs by providing explainability to the decision-making process. Supplementary Information The online version contains supplementary material available at 10.1186/s12859-022-04767-y.
| S-EPMC9204982 | biostudies-literature