Knowledge-Based Conformer Generation Using the Cambridge Structural Database.

Fast generation of plausible molecular conformations is central to molecular modeling. This paper presents an approach to conformer generation that makes extensive use of the information available in the Cambridge Structural Database. By using geometric distributions derived from the Cambridge Structural Database, it is possible to create biologically relevant conformations in the majority of cases analyzed. The paper compares the performance of the approach with previously published evaluations, and presents some cases where the method fails. The method appears to show significantly improved performance in reproduction of the conformations of structures observed in the Cambridge Structural Database and the Protein Data Bank as compared to other published methods of a similar speed.

Knowledge-Based Optimization of Molecular Geometries Using Crystal Structures.

This paper describes a novel way to use the structural information contained in the Cambridge Structural Database (CSD) to drive geometry optimization of organic molecules. We describe how CSD structural information is transformed into objective functions for gradient-based optimization to provide good quality geometries for a large variety of organic molecules. Performance is assessed by minimizing different sets of organic molecules reporting RMSD movements for bond lengths, valence angles, torsion angles, and heavy atom positions.

Kernel density estimation applied to bond length, bond angle, and torsion angle distributions.

We describe the method of kernel density estimation (KDE) and apply it to molecular structure data. KDE is a quite general nonparametric statistical method suitable even for multimodal data. The method generates smooth probability density function (PDF) representations and finds application in diverse fields such as signal processing and econometrics. KDE appears to have been under-utilized as a method in molecular geometry analysis, chemo-informatics, and molecular structure optimization. The resulting probability densities have advantages over histograms and, importantly, are also suitable for gradient-based optimization. To illustrate KDE, we describe its application to chemical bond length, bond valence angle, and torsion angle distributions and show the ability of the method to model arbitrary torsion angle distributions.

Knowledge-based libraries for predicting the geometric preferences of druglike molecules.

We describe the automated generation of libraries for predicting the geometric preferences of druglike molecules. The libraries contain distributions of molecular dimensions based on crystal structures in the Cambridge Structural Database (CSD). Searching of the libraries is performed in cascade fashion to identify the most relevant distributions in cases where precise structural features are poorly represented by existing crystal structures. The libraries are fully comprehensive for bond lengths, valence angles, and rotamers and produce templates for the large majority of unfused and fused rings. Geometry distributions for rotamers and rings take into account any atom chirality that may be present. Library validation has been performed on a set of druglike molecules whose structures were published after the latest CSD entry contributing to the libraries. Hence, the validation gives a true indication of prediction accuracy.

Potential and limitations of ensemble docking.

A major problem in structure-based virtual screening applications is the appropriate selection of a single or even multiple protein structures to be used in the virtual screening process. A priori it is unknown which protein structure(s) will perform best in a virtual screening experiment. We investigated the performance of ensemble docking, as a function of ensemble size, for eight targets of pharmaceutical interest. Starting from single protein structure docking results, for each ensemble size up to 500,000 combinations of protein structures were generated, and, for each ensemble, pose prediction and virtual screening results were derived. Comparison of single to multiple protein structure results suggests improvements when looking at the performance of the worst and the average over all single protein structures to the performance of the worst and average over all protein ensembles of size two or greater, respectively. We identified several key factors affecting ensemble docking performance, including the sampling accuracy of the docking algorithm, the choice of the scoring function, and the similarity of database ligands to the cocrystallized ligands of ligand-bound protein structures in an ensemble. Due to these factors, the prospective selection of optimum ensembles is a challenging task, shown by a reassessment of published ensemble selection protocols.

Pose prediction and virtual screening performance of GOLD scoring functions in a standardized test.

The performance of all four GOLD scoring functions has been evaluated for pose prediction and virtual screening under the standardized conditions of the comparative docking and scoring experiment reported in this Edition. Excellent pose prediction and good virtual screening performance was demonstrated using unmodified protein models and default parameter settings. The best performing scoring function for both pose prediction and virtual screening was demonstrated to be the recently introduced scoring function ChemPLP. We conclude that existing docking programs already perform close to optimally in the cognate pose prediction experiments currently carried out and that more stringent pose prediction tests should be used in the future. These should employ cross-docking sets. Evaluation of virtual screening performance remains problematic and much remains to be done to improve the usefulness of publically available active and decoy sets for virtual screening. Finally we suggest that, for certain target/scoring function combinations, good enrichment may sometimes be a consequence of 2D property recognition rather than a modelling of the correct 3D interactions.

Are predefined decoy sets of ligand poses able to quantify scoring function accuracy?

Due to the large number of different docking programs and scoring functions available, researchers are faced with the problem of selecting the most suitable one when starting a structure-based drug discovery project. To guide the decision process, several studies comparing different docking and scoring approaches have been published. In the context of comparing scoring function performance, it is common practice to use a predefined, computer-generated set of ligand poses (decoys) and to reevaluate their score using the set of scoring functions to be compared. But are predefined decoy sets able to unambiguously evaluate and rank different scoring functions with respect to pose prediction performance? This question arose when the pose prediction performance of our piecewise linear potential derived scoring functions (Korb et al. in J Chem Inf Model 49:84-96, 2009) was assessed on a standard decoy set (Cheng et al. in J Chem Inf Model 49:1079-1093, 2009). While they showed excellent pose identification performance when they were used for rescoring of the predefined decoy conformations, a pronounced degradation in performance could be observed when they were directly applied in docking calculations using the same test set. This implies that on a discrete set of ligand poses only the rescoring performance can be evaluated. For comparing the pose prediction performance in a more rigorous manner, the search space of each scoring function has to be sampled extensively as done in the docking calculations performed here. We were able to identify relative strengths and weaknesses of three scoring functions (ChemPLP, GoldScore, and Astex Statistical Potential) by analyzing the performance for subsets of the complexes grouped by different properties of the active site. However, reasons for the overall poor performance of all three functions on this test set compared to other test sets of similar size could not be identified.

Development and validation of an improved algorithm for overlaying flexible molecules.

A program for overlaying multiple flexible molecules has been developed. Candidate overlays are generated by a novel fingerprint algorithm, scored on three objective functions (union volume, hydrogen-bond match, and hydrophobic match), and ranked by constrained Pareto ranking. A diverse subset of the best ranked solutions is chosen using an overlay-dissimilarity metric. If necessary, the solutions can be optimised. A multi-objective genetic algorithm can be used to find additional overlays with a given mapping of chemical features but different ligand conformations. The fingerprint algorithm may also be used to produce constrained overlays, in which user-specified chemical groups are forced to be superimposed. The program has been tested on several sets of ligands, for each of which the true overlay is known from protein-ligand crystal structures. Both objective and subjective success criteria indicate that good results are obtained on the majority of these sets.

The ensemble performance index: an improved measure for assessing ensemble pose prediction performance.

We present a theoretical study on the performance of ensemble docking methodologies considering multiple protein structures. We perform a theoretical analysis of pose prediction experiments which is completely unbiased, as we make no assumptions about specific scoring functions, search paradigms, protein structures, or ligand data sets. We introduce a novel interpretable measure, the ensemble performance index (EPI), for the assessment of scoring performance in ensemble docking, which will be applied to simulated and real data sets.

Accelerating molecular docking calculations using graphics processing units.

The generation of molecular conformations and the evaluation of interaction potentials are common tasks in molecular modeling applications, particularly in protein-ligand or protein-protein docking programs. In this work, we present a GPU-accelerated approach capable of speeding up these tasks considerably. For the evaluation of interaction potentials in the context of rigid protein-protein docking, the GPU-accelerated approach reached speedup factors of up to over 50 compared to an optimized CPU-based implementation. Treating the ligand and donor groups in the protein binding site as flexible, speedup factors of up to 16 can be observed in the evaluation of protein-ligand interaction potentials. Additionally, we introduce a parallel version of our protein-ligand docking algorithm PLANTS that can take advantage of this GPU-accelerated scoring function evaluation. We compared the GPU-accelerated parallel version to the same algorithm running on the CPU and also to the highly optimized sequential CPU-based version. In terms of dependence of the ligand size and the number of rotatable bonds, speedup factors of up to 10 and 7, respectively, can be observed. Finally, a fitness landscape analysis in the context of rigid protein-protein docking was performed. Using a systematic grid-based search methodology, the GPU-accelerated version outperformed the CPU-based version with speedup factors of up to 60.

pharmACOphore: multiple flexible ligand alignment based on ant colony optimization.

The flexible superimposition of biologically active ligands is a crucial step in ligand-based drug design. Here we present pharmACOphore, a new approach for pairwise as well as multiple flexible alignment of ligands based on ant colony optimization (ACO; Dorigo, M.; Stützle, T. Ant Colony Optimization; MIT Press: Cambridge, MA, USA, 2004). An empirical scoring function is used, which describes ligand similarity by minimizing the distance of pharmacophoric features. The scoring function was parametrized on pairwise alignments of ligand sets for four proteins from diverse protein families (cyclooxygenase-2, cyclin-dependent kinase 2, factor Xa and peroxisome proliferator-activated receptor γ). The derived parameters were assessed with respect to pose prediction performance on the independent FlexS data set ( Lemmen, C.; Lengauer, T.; Klebe, G. J. Med. Chem. 1998, 41, 4502 - 4520) in exhausting pairwise alignments. Additionally, multiple flexible alignment experiments were carried out for the pharmacologically relevant targets trypsin and poly (ADP-ribose) polymerase (PARP). The results obtained show that the new procedure provides a robust and efficient way for the pairwise as well as multiple flexible alignment of small molecules.

How Significant Are Unusual Protein-Ligand Interactions? Insights from Database Mining.

We present a new approach to derive interaction propensities of protein-ligand atom pairs from mining of the Protein Data Bank. To ensure solid statistics, we use a line-of-sight contact filter and normalize the observed frequency of hits by a statistical null model based on exposed surface areas of atom types in the protein-ligand binding site. This allows us to investigate which intermolecular interactions and geometries are found more often than expected by chance in protein-ligand complexes. We focus our study on some of the unusual interactions that were postulated to be favorable, including σ-hole bonding of halogen and sulfur atoms, weak hydrogen bonding with fluorine as acceptor, and different types of dipolar interactions. Our results confirm some and challenge other common assumptions on these interactions and highlight other contact types that are yet underexplored in structure-based drug design.

Antagonists of the miRNA-Argonaute 2 Protein Complex: Anti-miR-AGOs.

microRNAs (miRNAs) have been identified as high-value drug targets. A widely applied strategy in miRNA inhibition is the use of antisense agents. However, it has been shown that oligonucleotides are poorly cell permeable because of their complex chemical structure and due to their negatively charged backbone. Consequently, the general application of oligonucleotides in therapy is limited. Since miRNAs' functions are executed exclusively by the Argonaute 2 protein, we therefore describe a protocol for the design of a novel miRNA inhibitor class: antagonists of the miRNA-Argonaute 2 protein complex, so-called anti-miR-AGOs, that not only block the crucial binding site of the target miRNA but also bind to the protein's active site. Due to their lower molecular weight and, thus, more drug-like chemical structure, the novel inhibitor class may show better pharmacokinetic properties than reported oligonucleotide inhibitors, enabling them for potential therapeutic use.

Interactive and Versatile Navigation of Structural Databases.

We present CSD-CrossMiner, a novel tool for pharmacophore-based searches in crystal structure databases. Intuitive pharmacophore queries describing, among others, protein-ligand interaction patterns, ligand scaffolds, or protein environments can be built and modified interactively. Matching crystal structures are overlaid onto the query and visualized as soon as they are available, enabling the researcher to quickly modify a hypothesis on the fly. We exemplify the utility of the approach by showing applications relevant to real-world drug discovery projects, including the identification of novel fragments for a specific protein environment or scaffold hopping. The ability to concurrently search protein-ligand binding sites extracted from the Protein Data Bank (PDB) and small organic molecules from the Cambridge Structural Database (CSD) using the same pharmacophore query further emphasizes the flexibility of CSD-CrossMiner. We believe that CSD-CrossMiner closes an important gap in mining structural data and will allow users to extract more value from the growing number of available crystal structures.

The cloud and other new computational methods to improve molecular modelling.

INTRODUCTION: Industrial, as well as academic, drug discovery efforts are usually supported by computational modelling techniques. Many of these techniques, such as virtual high-throughput docking, pharmacophore-based screening of conformer databases and molecular dynamics simulations, are computationally very demanding. Depending on the parallelisation strategy applicable to the respective method, recent technologies based on central processing units, for example, cloud and grid computing, or graphics processing units (GPUs), can be employed to accelerate their execution times considerably. This allows the molecular modeller to look at larger data sets, or to use more accurate methods. AREAS COVERED: The article introduces the recent developments in grid, cloud and GPU computing. The authors provide an overview of molecular modelling applications running on the above-mentioned hardware platforms and highlight caveats of the respective architectures, both from a theoretical and a practical point of view. EXPERT OPINION: The architectures described can improve the molecular modelling process considerably, if the appropriate technologies are selected for the respective application. Despite these improvements, each of the individual computational platforms suffers from specific issues, which will need to be addressed in the future. Furthermore, current endeavours have focused on improving the performance of existing algorithms, rather than the development of new methods that explicitly harness these new technologies.

MicroRNA-specific argonaute 2 protein inhibitors.

As microRNA silencing processes are mediated by the protein Argonaute 2 and for target RNA binding only a short sequence at the microRNA's 5' end (seed region) is crucial, we report a novel inhibitor class: the microRNA-specific Argonaute 2 protein inhibitors that not only block this short recognition sequence but also bind to the protein's active site. We developed a model for rational drug design, enabling the identification of Argonaute 2 active site binders and their linkage with a peptide nucleic acid sequence (PNA), which addresses the microRNA of interest. The designed inhibitors targeting microRNA-122, a hepatitis C virus drug target, had an IC50 of 100 nM, 10-fold more active than the simple PNA sequence (IC50 of 1 µM), giving evidence that the strategy has potential. Due to their lower molecular weight, these inhibitors may show better pharmacokinetic properties than reported oligonucleotide inhibitors, enabling them for potential therapeutic use.

Discovery of Schaeffer's acid analogues as lead structures of mycobacterium tuberculosis type†II dehydroquinase using a rational drug design approach.

Rational ligand design: Schaeffer's acid analogues were identified as novel inhibitors of M. tuberculosis type II dehydroquinase, a key enzyme of the shikimate pathway. Their likely binding mode was predicted using a combination of ensemble docking and flexible active site residues. Potentially, this scaffold could provide a good starting point for the design of antitubercular agents.

Design, synthesis, and biological evaluation of an allosteric inhibitor of HSET that targets cancer cells with supernumerary centrosomes.

Centrosomes associate with spindle poles; thus, the presence of two centrosomes promotes bipolar spindle assembly in normal cells. Cancer cells often contain supernumerary centrosomes, and to avoid multipolar mitosis and cell death, these are clustered into two poles by the microtubule motor protein HSET. We report the discovery of an allosteric inhibitor of HSET, CW069, which we designed using a methodology on an interface of chemistry and biology. Using this approach, we explored millions of compounds in silico and utilized convergent syntheses. Only compound CW069 showed marked activity against HSET in vitro. The inhibitor induced multipolar mitoses only in cells containing supernumerary centrosomes. CW069 therefore constitutes a valuable tool for probing HSET function and, by reducing the growth of cells containing supernumerary centrosomes, paves the way for new cancer therapeutics.

Docking performance of fragments and druglike compounds.

This paper addresses two questions of key interest to researchers working with protein-ligand docking methods: (i) Why is there such a large variation in docking performance between different test sets reported in the literature? (ii) Are fragments more difficult to dock than druglike compounds? To answer these, we construct a test set of in-house X-ray structures of protein-ligand complexes from drug discovery projects, half of which contain fragment ligands, the other half druglike ligands. We find that a key factor affecting docking performance is ligand efficiency (LE). High LE compounds are significantly easier to dock than low LE compounds, which we believe could explain the differences observed between test sets reported in the literature. There is no significant difference in docking performance between fragments and druglike compounds, but the reasons why dockings fail appear to be different.

Prediction of framework-guest systems using molecular docking.

Established methods for docking ligands into protein binding sites are used in a different way--docking guest molecules into crystalline channel-based host systems. This work shows that the principles of docking can be successfully applied to the crystal engineering problem of predicting host-guest systems.