A knowledge-based weighting approach to ligand-based virtual screening

On the basis of the recently introduced reduced graph concept of ErG (extending reduced graphs), a straightforward weighting approach to include additional (e.g., structural or SAR) knowledge into similarity searching procedures for virtual screening (wErG) is proposed. This simple procedure is exemplified with three data sets, for which interaction patterns available from X-ray structures of native or peptidomimetic ligands with their target protein are used to significantly improve retrieval rates of known actives from the MDL Drug Report database. The results are compared to those of other virtual screening techniques such as Daylight fingerprints, FTrees, UNITY, and various FlexX docking protocols. Here, it is shown that wErG exhibits a very good and stable performance independent of the target structure. On the basis of this (and the fact that ErG retrieves structurally more dissimilar compounds due to its potential to perform scaffold-hopping), the combination of wErG and FlexX is successfully explored. Overall, wErG is not only an easily applicable weighting procedure that efficiently identifies actives in large data sets but it is also straightforward to understand for both medicinal and computational chemists and can, therefore, be driven by several aspects of project-related knowledge (e.g., X-ray, NMR, SAR, and site-directed mutagenesis) in a very early stage of the hit identification process.     

New fragment weighting scheme for the Bayesian inference network in ligand-based virtual screening

Many of the conventional similarity methods assume that molecular fragments that do not relate to biological activity carry the same weight as the important ones. One possible approach to this problem is to use the Bayesian inference network (BIN), which models molecules and reference structures as probabilistic inference networks. The relationships between molecules and reference structures in the Bayesian network are encoded using a set of conditional probability distributions, which can be estimated by the fragment weighting function, a function of the frequencies of the fragments in the molecule or the reference structure as well as throughout the collection. The weighting function combines one or more fragment weighting schemes. In this paper, we have investigated five different weighting functions and present a new fragment weighting scheme. Later on, these functions were modified to combine the new weighting scheme. Simulated virtual screening experiments with the MDL Drug Data Report (23) and maximum unbiased validation data sets show that the use of new weighting scheme can provide significantly more effective screening when compared with the use of current weighting schemes.     

ROCS-derived features for virtual screening

Rapid overlay of chemical structures (ROCS) is a standard tool for the calculation of 3D shape and chemical (“color”) similarity. ROCS uses unweighted sums to combine many aspects of similarity, yielding parameter-free models for virtual screening. In this report, we decompose the ROCS color force field into color components and color atom overlaps, novel color similarity features that can be weighted in a system-specific manner by machine learning algorithms. In cross-validation experiments, these additional features significantly improve virtual screening performance relative to standard ROCS.     

Community benchmarks for virtual screening

Ligand enrichment among top-ranking hits is a key metric of virtual screening. To avoid bias, decoys should resemble ligands physically, so that enrichment is not attributable to simple differences of gross features. We therefore created a directory of useful decoys (DUD) by selecting decoys that resembled annotated ligands physically but not topologically to benchmark docking performance. DUD has 2950 annotated ligands and 95,316 property-matched decoys for 40 targets. It is by far the largest and most comprehensive public data set for benchmarking virtual screening programs that I am aware of. This paper outlines several ways that DUD can be improved to provide better telemetry to investigators seeking to understand both the strengths and the weaknesses of current docking methods. I also highlight several pitfalls for the unwary: a risk of over-optimization, questions about chemical space, and the proper scope for using DUD. Careful attention to both the composition of benchmarks and how they are used is essential to avoid being misled by overfitting and bias.     

Evaluations of molecular docking programs for virtual screening

Structure-based virtual screening is carried out using molecular docking programs. A number of such docking programs are currently available, and the selection of docking program is difficult without knowing the characteristics or performance of each program. In this study, the screening performances of three molecular docking programs, DOCK, AutoDock, and GOLD, were evaluated with 116 target proteins. The screening performances were validated using two novel standards, along with a traditional enrichment rate measurement. For the evaluations, each docking run was repeated 1000 times with three initial conformations of a ligand. While each docking program has some merit over the other docking programs in some aspects, DOCK showed an unexpectedly better screening performance in the enrichment rates. Finally, we made several recommendations based on the evaluation results to enhance the screening performances of the docking programs.     

Quantum probability ranking principle for ligand-based virtual screening

Chemical libraries contain thousands of compounds that need screening, which increases the need for computational methods that can rank or prioritize compounds. The tools of virtual screening are widely exploited to enhance the cost effectiveness of lead drug discovery programs by ranking chemical compounds databases in decreasing probability of biological activity based upon probability ranking principle (PRP). In this paper, we developed a novel ranking approach for molecular compounds inspired by quantum mechanics, called quantum probability ranking principle (QPRP). The QPRP ranking criteria would make an attempt to draw an analogy between the physical experiment and molecular structure ranking process for 2D fingerprints in ligand based virtual screening (LBVS). The development of QPRP criteria in LBVS has employed the concepts of quantum at three different levels, firstly at representation level, this model makes an effort to develop a new framework of molecular representation by connecting the molecular compounds with mathematical quantum space. Secondly, estimate the similarity between chemical libraries and references based on quantum-based similarity searching method. Finally, rank the molecules using QPRP approach. Simulated virtual screening experiments with MDL drug data report (MDDR) data sets showed that QPRP outperformed the classical ranking principle (PRP) for molecular chemical compounds.     

Congestion game scheduling for virtual drug screening optimization

In virtual drug screening, the chemical diversity of hits is an important factor, along with their predicted activity. Moreover, interim results are of interest for directing the further research, and their diversity is also desirable. In this paper, we consider a problem of obtaining a diverse set of virtual screening hits in a short time. To this end, we propose a mathematical model of task scheduling for virtual drug screening in high-performance computational systems as a congestion game between computational nodes to find the equilibrium solutions for best balancing the number of interim hits with their chemical diversity. The model considers the heterogeneous environment with workload uncertainty, processing time uncertainty, and limited knowledge about the input dataset structure. We perform computational experiments and evaluate the performance of the developed approach considering organic molecules database GDB-9. The used set of molecules is rich enough to demonstrate the feasibility and practicability of proposed solutions. We compare the algorithm with two known heuristics used in practice and observe that game-based scheduling outperforms them by the hit discovery rate and chemical diversity at earlier steps. Based on these results, we use a social utility metric for assessing the efficiency of our equilibrium solutions and show that they reach greatest values.     

Fine tuning for success in structure-based virtual screening

Journal of Computer-Aided Molecular Design - Structure-based virtual screening plays a significant role in drug-discovery. The method virtually docks millions of compounds from corporate or public...     

Supervised consensus scoring for docking and virtual screening

Docking programs are widely used to discover novel ligands efficiently and can predict protein-ligand complex structures with reasonable accuracy and speed. However, there is an emerging demand for better performance from the scoring methods. Consensus scoring (CS) methods improve the performance by compensating for the deficiencies of each scoring function. However, conventional CS and existing scoring functions have the same problems, such as a lack of protein flexibility, inadequate treatment of salvation, and the simplistic nature of the energy function used. Although there are many problems in current scoring functions, we focus our attention on the incorporation of unbound ligand conformations. To address this problem, we propose supervised consensus scoring (SCS), which takes into account protein-ligand binding process using unbound ligand conformations with supervised learning. An evaluation of docking accuracy for 100 diverse protein-ligand complexes shows that SCS outperforms both CS and 11 scoring functions (PLP, F-Score, LigScore, DrugScore, LUDI, X-Score, AutoDock, PMF, G-Score, ChemScore, and D-score). The success rates of SCS range from 89% to 91% in the range of rmsd < 2 A, while those of CS range from 80% to 85%, and those of the scoring functions range from 26% to 76%. Moreover, we also introduce a method for judging whether a compound is active or inactive with the appropriate criterion for virtual screening. SCS performs quite well in docking accuracy and is presumably useful for screening large-scale compound databases before predicting binding affinity.     

Collection and preparation of molecular databases for virtual screening

Drug discovery and development research is undergoing a paradigm shift from a linear and sequential nature of the various steps involved in the drug discovery process of the past to the more parallel approach of the present, due to a lack of sufficient correlation between activities estimated by in vitro and in vivo assays. This is attributed to the non-drug-likeness of the lead molecules, which has often been detected at advanced drug development stages. Thus a striking aspect of this paradigm shift has been early/parallel in silico prioritization of drug-like molecular databases (also database pre-processing), in addition to prioritizing compounds with high affinity and selectivity for a protein target. In view of this, a drug-like database useful for virtual screening has been created by prioritizing molecules from 36 catalog suppliers, using our recently derived binary QSAR based drug-likeness model as a filter. The performance of this model was assessed by a comparative evaluation with respect to commonly used filters implemented by the ZINC database. Since the model was derived considering all the limitations that have plagued the existing rules and models, it performs better than the existing filters and thus the molecules prioritized by this filter represent a better subset of drug-like compounds. The application of this model on exhaustive subsets of 4,972,123 molecules, many of which have passed the ZINC database filters for drug-likeness, led to a further prioritization of 2,920,551 drug-like molecules. This database may have a great potential for in silico virtual screening for discovering molecules, which may survive the later stages of the drug development research.     

Collection and preparation of molecular databases for virtual screening

Drug discovery and development research is undergoing a paradigm shift from a linear and sequential nature of the various steps involved in the drug discovery process of the past to the more parallel approach of the present, due to a lack of sufficient correlation between activities estimated by in vitro and in vivo assays. This is attributed to the non-drug-likeness of the lead molecules, which has often been detected at advanced drug development stages. Thus a striking aspect of this paradigm shift has been early/parallel in silico prioritization of drug-like molecular databases (also database pre-processing), in addition to prioritizing compounds with high affinity and selectivity for a protein target. In view of this, a drug-like database useful for virtual screening has been created by prioritizing molecules from 36 catalog suppliers, using our recently derived binary QSAR based drug-likeness model as a filter. The performance of this model was assessed by a comparative evaluation with respect to commonly used filters implemented by the ZINC database. Since the model was derived considering all the limitations that have plagued the existing rules and models, it performs better than the existing filters and thus the molecules prioritized by this filter represent a better subset of drug-like compounds. The application of this model on exhaustive subsets of 4,972,123 molecules, many of which have passed the ZINC database filters for drug-likeness, led to a further prioritization of 2,920,551 drug-like molecules. This database may have a great potential for in silico virtual screening for discovering molecules, which may survive the later stages of the drug development research.     

1001 Ways to run AutoDock Vina for virtual screening

Large-scale computing technologies have enabled high-throughput virtual screening involving thousands to millions of drug candidates. It is not trivial, however, for biochemical scientists to evaluate the technical alternatives and their implications for running such large experiments. Besides experience with the molecular docking tool itself, the scientist needs to learn how to run it on high-performance computing (HPC) infrastructures, and understand the impact of the choices made. Here, we review such considerations for a specific tool, AutoDock Vina, and use experimental data to illustrate the following points: (1) an additional level of parallelization increases virtual screening throughput on a multi-core machine; (2) capturing of the random seed is not enough (though necessary) for reproducibility on heterogeneous distributed computing systems; (3) the overall time spent on the screening of a ligand library can be improved by analysis of factors affecting execution time per ligand, including number of active torsions, heavy atoms and exhaustiveness. We also illustrate differences among four common HPC infrastructures: grid, Hadoop, small cluster and multi-core (virtual machine on the cloud). Our analysis shows that these platforms are suitable for screening experiments of different sizes. These considerations can guide scientists when choosing the best computing platform and set-up for their future large virtual screening experiments.     

Improved cyclodextrin-based receptors for camptothecin by inverse virtual screening

We report the computer-aided optimization of a synthetic receptor for a given guest molecule, based on inverse virtual screening of receptor libraries. As an example, a virtual set of beta-cyclodextrin (beta-CD) derivatives was generated as receptor candidates for the anticancer drug camptothecin. We applied the two docking tools AutoDock and GlamDock to generate camptothecin complexes of every candidate receptor. Scoring functions were used to rank all generated complexes. From the 10 % top-ranking candidates nine were selected for experimental validation. They were synthesized by reaction of heptakis-[6-deoxy-6-iodo]-beta-CD with a thiol compound to form the hepta-substituted beta-CDs. The stabilities of the camptothecin complexes obtained from solubility measurements of five of the nine CD derivatives were significantly higher than for any other CD derivative known from literature. The remaining four CD derivatives were insoluble in water. In addition, corresponding mono-substituted CD derivatives were synthesized that also showed improved binding constants. Among them the 9-H-purine derivative was the best, being comparable to the investigated hepta-substituted beta-CDs. Since the measured binding free energies correlated satisfactorily with the calculated scores, the applied scoring functions appeared to be appropriate for the selection of promising candidates for receptor synthesis.     

Consensus scoring criteria for improving enrichment in virtual screening

MOTIVATION: Virtual screening of molecular compound libraries is a potentially powerful and inexpensive method for the discovery of novel lead compounds for drug development. The major weakness of virtual screening-the inability to consistently identify true positives (leads)-is likely due to our incomplete understanding of the chemistry involved in ligand binding and the subsequently imprecise scoring algorithms. It has been demonstrated that combining multiple scoring functions (consensus scoring) improves the enrichment of true positives. Previous efforts at consensus scoring have largely focused on empirical results, but they have yet to provide a theoretical analysis that gives insight into real features of combinations and data fusion for virtual screening. RESULTS: We demonstrate that combining multiple scoring functions improves the enrichment of true positives only if (a) each of the individual scoring functions has relatively high performance and (b) the individual scoring functions are distinctive. Notably, these two prediction variables are previously established criteria for the performance of data fusion approaches using either rank or score combinations. This work, thus, establishes a potential theoretical basis for the probable success of data fusion approaches to improve yields in in silico screening experiments. Furthermore, it is similarly established that the second criterion (b) can, in at least some cases, be functionally defined as the area between the rank versus score plots generated by the two (or more) algorithms. Because rank-score plots are independent of the performance of the individual scoring function, this establishes a second theoretically defined approach to determining the likely success of combining data from different predictive algorithms. This approach is, thus, useful in practical settings in the virtual screening process when the performance of at least two individual scoring functions (such as in criterion a) can be estimated as having a high likelihood of having high performance, even if no training sets are available. We provide initial validation of this theoretical approach using data from five scoring systems with two evolutionary docking algorithms on four targets, thymidine kinase, human dihydrofolate reductase, and estrogen receptors of antagonists and agonists. Our procedure is computationally efficient, able to adapt to different situations, and scalable to a large number of compounds as well as to a greater number of combinations. Results of the experiment show a fairly significant improvement (vs single algorithms) in several measures of scoring quality, specifically "goodness-of-hit" scores, false positive rates, and "enrichment". This approach (available online at http://gemdock.life. nctu.edu.tw/dock/download.php) has practical utility for cases where the basic tools are known or believed to be generally applicable, but where specific training sets are absent.     

Consensus scoring with feature selection for structure-based virtual screening

The evaluation of ligand conformations is a crucial aspect of structure-based virtual screening, and scoring functions play significant roles in it. While consensus scoring (CS) generally improves enrichment by compensating for the deficiencies of each scoring function, the strategy of how individual scoring functions are selected remains a challenging task when few known active compounds are available. To address this problem, we propose feature selection-based consensus scoring (FSCS), which performs supervised feature selection with docked native ligand conformations to select complementary scoring functions. We evaluated the enrichments of five scoring functions (F-Score, D-Score, PMF, G-Score, and ChemScore), FSCS, and RCS (rank-by-rank consensus scoring) for four different target proteins: acetylcholine esterase (AChE), thrombin (thrombin), phosphodiesterase 5 (PDE5), and peroxisome proliferator-activated receptor gamma (PPARgamma). The results indicated that FSCS was able to select the complementary scoring functions and enhance ligand enrichments and that it outperformed RCS and the individual scoring functions for all target proteins. They also indicated that the performances of the single scoring functions were strongly dependent on the target protein. An especially favorable result with implications for practical drug screening is that FSCS performs well even if only one 3D structure of the protein-ligand complex is known. Moreover, we found that one can infer which scoring functions significantly enrich active compounds by using feature selection before actual docking and that the selected scoring functions are complementary.     

StructRank: a new approach for ligand-based virtual screening

Screening large libraries of chemical compounds against a biological target, typically a receptor or an enzyme, is a crucial step in the process of drug discovery. Virtual screening (VS) can be seen as a ranking problem which prefers as many actives as possible at the top of the ranking. As a standard, current Quantitative Structure-Activity Relationship (QSAR) models apply regression methods to predict the level of activity for each molecule and then sort them to establish the ranking. In this paper, we propose a top-k ranking algorithm (StructRank) based on Support Vector Machines to solve the early recognition problem directly. Empirically, we show that our ranking approach outperforms not only regression methods but another ranking approach recently proposed for QSAR ranking, RankSVM, in terms of actives found.