Maximum Common Binding Modes (MCBM): Consensus Docking Scoring Using Multiple Ligand Information and Interaction Fingerprints

Leveraging nonstructural data to predict structures and affinities of protein-ligand complexes

Structure-based drug design depends on the ability to predict both the three-dimensional structures of candidate molecules bound to their targets and the associated binding affinities. We demonstrate that one can substantially improve the accuracy of these predictions using easily obtained data about completely different molecules that bind to the same target without requiring any target-bound structures of these molecules. The approach we developed to integrate physical and data-driven modeling may find a variety of applications in the rapidly growing field of artificial intelligence for drug discovery.

Over the past five decades, tremendous effort has been devoted to computational methods for predicting properties of ligands—i.e., molecules that bind macromolecular targets. Such methods, which are critical to rational drug design, fall into two categories: physics-based methods, which directly model ligand interactions with the target given the target’s three-dimensional (3D) structure, and ligand-based methods, which predict ligand properties given experimental measurements for similar ligands. Here, we present a rigorous statistical framework to combine these two sources of information. We develop a method to predict a ligand’s pose—the 3D structure of the ligand bound to its target—that leverages a widely available source of information: a list of other ligands that are known to bind the same target but for which no 3D structure is available. This combination of physics-based and ligand-based modeling improves pose prediction accuracy across all major families of drug targets. Using the same framework, we develop a method for virtual screening of drug candidates, which outperforms standard physics-based and ligand-based virtual screening methods. Our results suggest broad opportunities to improve prediction of various ligand properties by combining diverse sources of information through customized machine-learning approaches.

Effective prediction model and determination of binding residues influential for inhibitors targeting HIV-1 integrase-LEDGF/p75 interface by employing solvent accessible surface area energy as key determinant

Development of a highly accurate prediction model for protein-ligand inhibition has been a major challenge in drug discovery. Herein, we describe a novel predictive model for the inhibition of HIV-1 integrase (IN)-LEDGF/p75 protein-protein interaction. The model was constructed using energy parameters approximated from molecular dynamics (MD) simulations and molecular mechanics Poisson-Boltzmann surface area (MM-PBSA) calculations. Chemometric analysis using partial least squares (PLS) regression revealed that solvent accessible surface area energy (ΔG_SASA) is the major determinant parameter contributing greatly to the prediction accuracy. PLS prediction model on the ΔG_SASA values collected from 41 complexes yielded a strong correlation between the predicted and the actual inhibitory activities (R² = 0.9666, RMSEC of pIC₅₀ values = 0.0890). Additionally, for the test set of 14 complexes, the model performed satisfactorily with very low pIC₅₀ errors (Q² = 0.5168, RMSEP = 0.3325). A strong correlation between the buried surface areas on the IN protein, when bound with IN-LEDGF/p75 inhibitors, and the respective ΔG_SASA values was also obtained. Furthermore, the current method could identify 'hot spots'of amino acid residues highly influential to the inhibitory activity prediction. This could present fruitful implications in binding site determination and future inhibitor developments targeting protein-protein interactions.Communicated by Ramaswamy H. Sarma.

COX Inhibition Profiles and Molecular Docking Studies of the Lignan Hinokinin and Some Synthetic Derivatives

Encouraged by the anti-inflammatory activity of hinokinin in vivo, which is also observed for the analogues dinitrohinokinin and diidrocubebin, herein we used in vitro and in silico methods to assess their selectivity profiles and predict their binding modes with Cyclooxygenases (COX-1 and 2). The in vitro assays demonstrated dinitrohinokinin is about 13 times more selective for COX-2 than for COX-1, a similar profile observed for the drugs celecoxib (selective index ≈9) and meloxicam (selective index ≈11). Predictions of the binding modes suggested dinitrohinokinin interacts with COX-2 very similarly to rofecoxib, exploring residues at the hydrophilic pocket of the enzyme that accessible to ligands only in this isoform. This lignan also interacts with COX-1 in a similar mode to meloxicam, blocking the access of the substrate to the catalytic cleft. Therefore, dinitrohinokinin is a promising lead for the design of selective COX-2 inhibitors.

Wogonin Suppresses the Activity of Matrix Metalloproteinase-9 and Inhibits Migration and Invasion in Human Hepatocellular Carcinoma

As one of the major active ingredients in Radix Scutellariae, wogonin has been shown to be associated with various pharmacological activities on cancer cell growth, apoptosis, and cell invasion and migration. Here, we demonstrated that wogonin may harbor potential anti-metastatic activities in hepatocarcinoma (HCC). The anti-metastasis potential of wogonin and its underlying mechanisms were evaluated by ligand–protein docking approach, surface plasmon resonance assay, and in vitro gelatin zymography studies. Our results showed that wogonin (100 μM, 50 μM) suppressed MHCC97L and PLC/PRF/5 cells migration and invasion in vitro. The docking approach and surface plasmon resonance assay indicated that the potential binding affinity between wogonin and matrix metalloproteinase-9 (MMP-9) may lead to inhibition of MMP-9 activity and further leads to suppression of tumor metastasis. This conclusion was further verified by Western blot results and gelatin zymography analysis. Wogonin might be a potent treatment option for disrupting the tumor metastasis that favors HCC development. The potential active targets from computational screening integrated with biomedical study may help us to explore the molecular mechanism of herbal medicines.

A Network Pharmacology-Based Study on the Hepatoprotective Effect of Fructus Schisandrae

Fructus schisandrae (Wuweizi in Chinese), a common traditional Chinese herbal medicine, has been used for centuries to treat chronic liver disease. The therapeutic efficacy of Wuweizi has also been validated in clinical practice. In this study, molecular docking and network analysis were carried out to explore the hepatoprotective mechanism of Wuweizi as an effective therapeutic approach to treat liver disease. Multiple active compounds of Wuweizi were docked with 44 protein targets related with viral hepatitis, fatty liver, liver fibrosis, cirrhosis, and liver cancer. A compound-target network was constructed through network pharmacology analysis, predicting the relationships of active ingredients to the targets. Our results demonstrated that schisantherin, schisandrin B, schisandrol B, kadsurin, Wuweizisu C, Gomisin A, Gomisin G, and angeloylgomisin may target with 21 intracellular proteins associated with liver diseases, especially with fatty liver disease. The CYP2E1, PPARα, and AMPK genes and their related pathway may play a pivotal role in the hepatoprotective effects of Wuweizi. The network pharmacology strategy used provides a forceful tool for searching the action mechanism of traditional herbal medicines and novel bioactive ingredients.

Identification of a novel putative inhibitor of the Plasmodium falciparum purine nucleoside phosphorylase: exploring the purine salvage pathway to design new antimalarial drugs

Malaria, a tropical parasitic disease caused by Plasmodium spp., continues to place a heavy social burden, with almost 200 million cases and more than 580,000 deaths per year. Plasmodium falciparum purine nucleoside phosphorylase (PfPNP) can be targeted for antimalarial drug design since its inhibition kills malaria parasites both in vitro and in vivo. Although the currently known inhibitors of PfPNP, immucillins, are orally available and of low toxicity to animals and humans, to the best of our knowledge, none of these compounds has entered clinical trials for the treatment of malaria. Using a pharmacophore-based virtual screening coupled to a consensual molecular docking approach, we identified 59 potential PfPNP inhibitors that are predicted to be orally absorbed in a Caco-2 cell model. Although most of these compounds are predicted to have high plasma protein binding levels, poor water solubility (except for compound 25) and CYP3A4 metabolic stability (except for 4, 7 and 8), four structures (4, 7, 8 and 25) remain as potential leads because of their plausible interaction with a specific hydrophobic pocket of PfPNP, which would confer them higher selectivity for PfPNP over human PNP. Additionally, both predicted Gibbs free energies for binding and molecular dynamics suggest that compound 4 may form a more stable complex with PfPNP than 5[Formula: see text]-methylthio-immucillin-H, a potent and selective inhibitor of PfPNP.

Protein Ligand Interaction Fingerprints

A most challenging part in docking-based virtual screening is the scoring functions implemented in various docking programs in order to evaluate different poses of the ligands inside the binding cavity of the receptor. Precise and trustable measurement of ligand-protein affinity for Structure-Based Virtual Screening (SB-VS) is therefore, an outstanding problem in docking studies. Empirical post-docking filters can be helpful as a way to provide various types of structure-activity information. Different types of interaction have been presented between the ligands and the receptor so far. Based on the diversity and importance of PLIF methods, this chapter will focus on the comparison of different protocols. The advantages and disadvantages of all methods will be discussed explicitly in this chapter as well as future sights for further progress in this field. Different classifications approaches for the protein-ligand interaction fingerprints were also discussed in this chapter.

QSAR modeling and molecular interaction analysis of natural compounds as potent neuraminidase inhibitors

The constructed SRA, HQSAR, almond and CoMSIA models have good predictive capability, which can evaluate and screen new compounds.

HybridDock: A Hybrid Protein-Ligand Docking Protocol Integrating Protein- and Ligand-Based Approaches

Structure-based molecular docking and ligand-based similarity search are two commonly used computational methods in computer-aided drug design. Structure-based docking tries to utilize the structural information on a drug target like protein, and ligand-based screening takes advantage of the information on known ligands for a target. Given their different advantages, it would be desirable to use both protein- and ligand-based approaches in drug discovery when information for both the protein and known ligands is available. Here, we have presented a general hybrid docking protocol, referred to as HybridDock, to utilize both the protein structures and known ligands by combining the molecular docking program MDock and the ligand-based similarity search method SHAFTS, and evaluated our hybrid docking protocol on the CSAR 2013 and 2014 exercises. The results showed that overall our hybrid docking protocol significantly improved the performance in both binding affinity and binding mode predictions, compared to the sole MDock program. The efficacy of the hybrid docking protocol was further confirmed using the combination of DOCK and SHAFTS, suggesting an alternative docking approach for modern drug design/discovery.

Exploring the potential of protein-based pharmacophore models in ligand pose prediction and ranking

Protein-based pharmacophore models derived from protein binding site atoms without the inclusion of any ligand information have become more popular in virtual screening studies. However, the accuracy of protein-based pharmacophore models for reproducing the critical protein-ligand interactions has never been explicitly assessed. In this study, we used known protein-ligand contacts from a large set of experimentally determined protein-ligand complexes to assess the quality of the protein-based pharmacophores in reproducing these critical contacts. We demonstrate how these contacts can be used to optimize the pharmacophore generation procedure to produce pharmacophore models that optimally cover the known protein-ligand interactions. Finally, we explored the potential of the optimized protein-based pharmacophore models for pose prediction and pose rankings. Our results demonstrate that there are significant variations in the success of protein-based pharmacophore models to reproduce native contacts and consequently native ligand poses dependent on the details of the pharmacophore generation process. We show that the generation of optimized protein-based pharmacophore models is a promising approach for ligand pose prediction and pose rankings.

Contact-based ligand-clustering approach for the identification of active compounds in virtual screening

Evaluation of docking results is one of the most important problems for virtual screening and in silico drug design. Modern approaches for the identification of active compounds in a large data set of docked molecules use energy scoring functions. One of the general and most significant limitations of these methods relates to inaccurate binding energy estimation, which results in false scoring of docked compounds. Automatic analysis of poses using self-organizing maps (AuPosSOM) represents an alternative approach for the evaluation of docking results based on the clustering of compounds by the similarity of their contacts with the receptor. A scoring function was developed for the identification of the active compounds in the AuPosSOM clustered dataset. In addition, the AuPosSOM efficiency for the clustering of compounds and the identification of key contacts considered as important for its activity, were also improved. Benchmark tests for several targets revealed that together with the developed scoring function, AuPosSOM represents a good alternative to the energy-based scoring functions for the evaluation of docking results.

A two-step target binding and selectivity support vector machines approach for virtual screening of dopamine receptor subtype-selective ligands

Target selective drugs, such as dopamine receptor (DR) subtype selective ligands, are developed for enhanced therapeutics and reduced side effects. In silico methods have been explored for searching DR selective ligands, but encountered difficulties associated with high subtype similarity and ligand structural diversity. Machine learning methods have shown promising potential in searching target selective compounds. Their target selective capability can be further enhanced. In this work, we introduced a new two-step support vector machines target-binding and selectivity screening method for searching DR subtype-selective ligands, which was tested together with three previously-used machine learning methods for searching D1, D2, D3 and D4 selective ligands. It correctly identified 50.6%–88.0% of the 21–408 subtype selective and 71.7%–81.0% of the 39–147 multi-subtype ligands. Its subtype selective ligand identification rates are significantly better than, and its multi-subtype ligand identification rates are comparable to the best rates of the previously used methods. Our method produced low false-hit rates in screening 13.56 M PubChem, 168,016 MDDR and 657,736 ChEMBLdb compounds. Molecular features important for subtype selectivity were extracted by using the recursive feature elimination feature selection method. These features are consistent with literature-reported features. Our method showed similar performance in searching estrogen receptor subtype selective ligands. Our study demonstrated the usefulness of the two-step target binding and selectivity screening method in searching subtype selective ligands from large compound libraries.

Protein binding site analysis by means of structural interaction fingerprint patterns

We introduce a new approach to the known concept of interaction profiles, based on Structural Interaction Fingerprints (SIFt), for precise and rapid binding site description. A set of scripts for batch generation and analysis of SIFt were prepared, and the implementation is computationally efficient and supports parallelization. It is based on a 9-digit binary interaction pattern that describes physical ligand-protein interactions in structures and models of ligand-protein complexes. The tool performs analysis and identifies binding site residues (crucial and auxiliary) and classifies interactions according to type (hydrophobic, aromatic, charge, polar, side chain, and backbone). It is convenient and easy to use, and gives manageable output data for both, interpretation and further processing. In the presented Letter, SIFts are applied to analyze binding sites in models of antagonist-5-HT7 receptor complexes and structures of cyclin dependent kinase 2-ligand complexes.

CSAR benchmark exercise of 2010: combined evaluation across all submitted scoring functions

As part of the Community Structure-Activity Resource (CSAR) center, a set of 343 high-quality, protein–ligand crystal structures were assembled with experimentally determined K_d or K_i information from the literature. We encouraged the community to score the crystallographic poses of the complexes by any method of their choice. The goal of the exercise was to (1) evaluate the current ability of the field to predict activity from structure and (2) investigate the properties of the complexes and methods that appear to hinder scoring. A total of 19 different methods were submitted with numerous parameter variations for a total of 64 sets of scores from 16 participating groups. Linear regression and nonparametric tests were used to correlate scores to the experimental values. Correlation to experiment for the various methods ranged R² = 0.58–0.12, Spearman ρ = 0.74–0.37, Kendall τ = 0.55–0.25, and median unsigned error = 1.00–1.68 pK_d units. All types of scoring functions—force field based, knowledge based, and empirical—had examples with high and low correlation, showing no bias/advantage for any particular approach. The data across all the participants were combined to identify 63 complexes that were poorly scored across the majority of the scoring methods and 123 complexes that were scored well across the majority. The two sets were compared using a Wilcoxon rank-sum test to assess any significant difference in the distributions of >400 physicochemical properties of the ligands and the proteins. Poorly scored complexes were found to have ligands that were the same size as those in well-scored complexes, but hydrogen bonding and torsional strain were significantly different. These comparisons point to a need for CSAR to develop data sets of congeneric series with a range of hydrogen-bonding and hydrophobic characteristics and a range of rotatable bonds.

Implementation and evaluation of a docking-rescoring method using molecular footprint comparisons

A docking-rescoring method, based on per-residue van der Waals (VDW), electrostatic (ES), or hydrogen bond (HB) energies has been developed to aid discovery of ligands that have interaction signatures with a target (footprints) similar to that of a reference. Biologically useful references could include known drugs, inhibitors, substrates, transition states, or side-chains that mediate protein-protein interactions. Termed footprint similarity (FPS) score, the method, as implemented in the program DOCK, was validated and characterized using: (1) pose identification, (2) crossdocking, (3) enrichment, and (4) virtual screening. Improvements in pose identification (6–12%) were obtained using footprint-based (FPS(VDW+ES)) vs. standard DOCK (DCE(VDW+ES)) scoring as evaluated on three large datasets (680–775 systems) from the SB2010 database. Enhanced pose identification was also observed using FPS (45.4% or 70.9%) compared with DCE (17.8%) methods to rank challenging crossdocking ensembles from carbonic anhydrase. Enrichment tests, for three representative systems, revealed FPSVDW+ES scoring yields significant early fold enrichment in the top 10% of ranked databases. For EGFR, top FPS poses are nicely accommodated in the molecular envelope defined by the reference in comparison with DCE, which yields distinct molecular weight bias toward larger molecules. Results from a representative virtual screen of ca. 1 million compounds additionally illustrate how ligands with footprints similar to a known inhibitor can readily be identified from within large commercially available databases. By providing an alternative way to rank ligand poses in a simple yet directed manner we anticipate that FPS scoring will be a useful tool for docking and structure-based design.

Recent trends and observations in the design of high-quality screening collections

The design of a high-quality screening collection is of utmost importance for the early drug-discovery process and provides, in combination with high-quality assay systems, the foundation of future discoveries. Herein, we review recent trends and observations to successfully expand the access to bioactive chemical space, including the feedback from hit assessment interviews of high-throughput screening campaigns; recent successes with chemogenomics target family approaches, the identification of new relevant target/domain families, diversity-oriented synthesis and new emerging compound classes, and non-classical approaches, such as fragment-based screening and DNA-encoded chemical libraries. The role of in silico library design approaches are emphasized.

[Molecular docking: role of intermolecular contacts in formation of complexes of proteins with nucleotides and peptides]

Knowledge of 3D-structure of protein-ligand complex is a major prerequisite for understanding the functioning mechanism of cellular proteins and membrane receptors. This is also of a great help in rational drug design projects. In the present paper we briefly review the molecular docking approaches used to predict possible orientation of a ligand in the protein binding site. The recent trends to improve the accuracy and efficiency of docking algorithms are demonstrated with the results obtained in Laboratory of Biomolecular Modeling. Particular attention is paid to protein-ligand hydrophobic and stacking interactions responsible for molecular recognition of ligand fragments. Such type of interactions are not always adequately represented in scoring criteria of docking applications that leads to mismatch in 3D-structure complexes predictions. That is why further inquiry of methods to account for these interactions is now the area of active research.

Virtual screening of selective multitarget kinase inhibitors by combinatorial support vector machines

Multitarget agents have been increasingly explored for enhancing efficacy and reducing countertarget activities and toxicities. Efficient virtual screening (VS) tools for searching selective multitarget agents are desired. Combinatorial support vector machines (C-SVM) were tested as VS tools for searching dual-inhibitors of 11 combinations of 9 anticancer kinase targets (EGFR, VEGFR, PDGFR, Src, FGFR, Lck, CDK1, CDK2, GSK3). C-SVM trained on 233-1,316 non-dual-inhibitors correctly identified 26.8%-57.3% (majority >36%) of the 56-230 intra-kinase-group dual-inhibitors (equivalent to the 50-70% yields of two independent individual target VS tools), and 12.2% of the 41 inter-kinase-group dual-inhibitors. C-SVM were fairly selective in misidentifying as dual-inhibitors 3.7%-48.1% (majority <20%) of the 233-1,316 non-dual-inhibitors of the same kinase pairs and 0.98%-4.77% of the 3,971-5,180 inhibitors of other kinases. C-SVM produced low false-hit rates in misidentifying as dual-inhibitors 1,746-4,817 (0.013%-0.036%) of the 13.56 M PubChem compounds, 12-175 (0.007%-0.104%) of the 168 K MDDR compounds, and 0-84 (0.0%-2.9%) of the 19,495-38,483 MDDR compounds similar to the known dual-inhibitors. C-SVM was compared to other VS methods Surflex-Dock, DOCK Blaster, kNN and PNN against the same sets of kinase inhibitors and the full set or subset of the 1.02 M Zinc clean-leads data set. C-SVM produced comparable dual-inhibitor yields, slightly better false-hit rates for kinase inhibitors, and significantly lower false-hit rates for the Zinc clean-leads data set. Combinatorial SVM showed promising potential for searching selective multitarget agents against intra-kinase-group kinases without explicit knowledge of multitarget agents.

In-silico approaches to multi-target drug discovery : computer aided multi-target drug design, multi-target virtual screening

Multi-target drugs against selective multiple targets improve therapeutic efficacy, safety and resistance profiles by collective regulations of a primary therapeutic target together with compensatory elements and resistance activities. Efforts have been made to employ in-silico methods for facilitating the search and design of selective multi-target agents. These methods have shown promising potential in facilitating drug discovery directed at selective multiple targets.

Automated site preparation in physics-based rescoring of receptor ligand complexes

Hydrogen atoms are not typically observable in X-ray crystal structures, but inferring their locations is often important in structure-based drug design. In addition, protonation states of the protein can change in response to ligand binding, as can the orientations of OH groups, a subtle form of "induced fit." We implement and evaluate an automated procedure for optimizing polar hydrogens in protein-binding sites in complex with ligands. Specifically, we apply the previously described Independent Cluster Decomposition Algorithm (ICDA) algorithm (Li et al., Proteins 2007;66:824-837), which assigns the ionization states of titratable residues, the amide orientations of Asn/Gln side chains, the imidazole ring orientation in His, and the orientations of OH/SH groups, in a unified algorithm. We test the utility of this method for identifying nativelike ligand poses using 247 protein-ligand complexes from an established database of docked decoys. Pose selection is performed with a physics-based scoring function based on a molecular mechanics energy function and a Generalized Born implicit solvent model. The use of the ICDA receptor preparation protocol, implemented with no knowledge of the native ligand pose, increases the accuracy of pose selection significantly, with the average RMSD over all complexes decreasing from 2.7 to 1.5 A when applying ICDA. Large improvements are seen for specific classes of binding sites with titratable groups, such as aspartyl proteases.

Docking ligands into flexible and solvated macromolecules. 4. Are popular scoring functions accurate for this class of proteins?

In our previous report, we investigated the impact of protein flexibility and the presence of water molecules on the pose-prediction accuracy of major docking programs. To complete these investigations, we report herein a study of the impact of these two aspects on the accuracy of scoring functions. To this effect, we developed two sets of protein/ligand complexes made up of ligands cross-docked or cocrystallized with a large variety of proteins, featuring bridging water molecules and demonstrating protein flexibility. Efforts were made to reduce the correlation between the molecular weights of the selected ligands and their binding affinities, a major bias in some previously reported benchmark sets. Using these sets, 18 available scoring functions have been assessed for their accuracy to predict binding affinities and to rank-order compounds by their affinity to cocrystallized proteins. This study confirmed the good and similar accuracy of Xscore, GlideScore, DrugScore(CSD), GoldScore, PLP1, ChemScore, RankScore, and the eHiTS scoring function. Our next investigations demonstrated that most of the assessed scoring functions were much less accurate when the correct protein conformation was not provided. This study also revealed that considering the water molecules for scoring does not greatly affect the accuracy. Finally, this work sheds light on the high correlation between scoring functions and the poor increase in accuracy one can expect from consensus scoring.

In pursuit of virtual lead optimization: pruning ensembles of receptor structures for increased efficiency and accuracy during docking

Representing receptors as ensembles of protein conformations during docking is a powerful method to approximate protein flexibility and increase the accuracy of the resulting ranked list of compounds. Unfortunately, docking compounds against a large number of ensemble members can increase computational cost and time investment. In this article, we present an efficient method to evaluate and select the most contributive ensemble members prior to docking for targets with a conserved core of residues that bind a ligand moiety. We observed that ensemble members that preserve the geometry of the active site core are most likely to place ligands in the active site with a conserved orientation, generally rank ligands correctly and increase interactions with the receptor. A relative distance approach is used to quantify the preservation of the three-dimensional interatomic distances of the conserved ligand-binding atoms and prune large ensembles quickly. In this study, we investigate dihydrofolate reductase as an example of a protein with a conserved core; however, this method for accurately selecting relevant ensemble members a priori can be applied to any system with a conserved ligand-binding core, including HIV-1 protease, kinases, and acetylcholinesterase. Representing a drug target as a pruned ensemble during in silico screening should increase the accuracy and efficiency of high-throughput analyses of lead analogs.