Benchmarking ligand-based virtual High-Throughput Screening with the PubChem database

Advancements in Ligand-Based Virtual Screening through the Synergistic Integration of Graph Neural Networks and Expert-Crafted Descriptors

The fusion of traditional chemical descriptors with Graph Neural Networks (GNNs) offers a compelling strategy for enhancing ligand-based virtual screening methodologies. A comprehensive evaluation revealed that the benefits derived from this integrative strategy vary significantly among different GNNs. Specifically, while GCN and SchNet demonstrate pronounced improvements by incorporating descriptors, SphereNet exhibits only marginal enhancement. Intriguingly, despite SphereNet's modest gain, all three models-GCN, SchNet, and SphereNet-achieve comparable performance levels when leveraging this combination strategy. This observation underscores a pivotal insight: sophisticated GNN architectures may be substituted with simpler counterparts without sacrificing efficacy, provided that they are augmented with descriptors. Furthermore, our analysis reveals a set of expert-crafted descriptors' robustness in scaffold-split scenarios, frequently outperforming the combined GNN-descriptor models. Given the critical importance of scaffold splitting in accurately mimicking real-world drug discovery contexts, this finding accentuates an imperative for GNN researchers to innovate models that can adeptly navigate and predict within such frameworks. Our work not only validates the potential of integrating descriptors with GNNs in advancing ligand-based virtual screening but also illuminates pathways for future enhancements in model development and application. Our implementation can be found at https://github.com/meilerlab/gnn-descriptor.

The Diagnostic Features of Peripheral Blood Biomarkers in Identifying Osteoarthritis Individuals: Machine Learning Strategies and Clinical Evidence

BACKGROUND

People with osteoarthritis place a huge burden on society. Early diagnosis is essential to prevent disease progression and to select the best treatment strategy more effectively. In this study, the aim was to examine the diagnostic features and clinical value of peripheral blood biomarkers for osteoarthritis.

OBJECTIVE

The goal of this project was to investigate the diagnostic features of peripheral blood and immune cell infiltration in osteoarthritis (OA).

METHODS

Two eligible datasets (GSE63359 and GSE48556) were obtained from the GEO database to discern differentially expressed genes (DEGs). The machine learning strategy was employed to filtrate diagnostic biomarkers for OA. Additional verification was implemented by collecting clinical samples of OA. The CIBERSORT website estimated relative subsets of RNA transcripts to evaluate the immune-inflammatory states of OA. The link between specific DEGs and clinical immune-inflammatory markers was found by correlation analysis.

RESULTS

Overall, 67 robust DEGs were identified. The nuclear receptor subfamily 2 group C member 2 (NR2C2), transcription factor 4 (TCF4), stromal antigen 1 (STAG1), and interleukin 18 receptor accessory protein (IL18RAP) were identified as effective diagnostic markers of OA in peripheral blood. All four diagnostic markers showed significant increases in expression in OA. Analysis of immune cell infiltration revealed that macrophages are involved in the occurrence of OA. Candidate diagnostic markers were correlated with clinical immune-inflammatory indicators of OA patients.

CONCLUSION

We highlight that DEGs associated with immune inflammation (NR2C2, TCF4, STAG1, and IL18RAP) may be potential biomarkers for peripheral blood in OA, which are also associated with clinical immune-inflammatory indicators.

Quorum Quenchers from Reynoutria japonica in the Battle against Methicillin-Resistant Staphylococcus aureus (MRSA)

Over the past decade, methicillin-resistant Staphylococcus aureus (MRSA) has become a major source of biofilm formation and a major contributor to antimicrobial resistance. The genes that govern biofilm formation are regulated by a signaling mechanism called the quorum-sensing system. There is a need for new molecules to treat the infections caused by dangerous pathogens like MRSA. The current study focused on an alternative approach using juglone derivatives from Reynoutria japonica as quorum quenchers. Ten bioactive compounds from this plant, i.e., 2-methoxy-6-acetyl-7-methyljuglone, emodin, emodin 8-o-b glucoside, polydatin, resveratrol, physcion, citreorosein, quercetin, hyperoside, and coumarin were taken as ligands and docked with accessory gene regulator proteins A, B, and C and the signal transduction protein TRAP. The best ligand was selected based on docking score, ADMET properties, and the Lipinski rule. Considering all these parameters, resveratrol displayed all required drug-like properties with a docking score of −8.9 against accessory gene regulator protein C. To further assess the effectiveness of resveratrol, it was compared with the commercially available antibiotic drug penicillin. A comparison of all drug-like characteristics showed that resveratrol was superior to penicillin in many aspects. Penicillin showed a binding affinity of −6.7 while resveratrol had a score of −8.9 during docking. This was followed by molecular dynamic simulations wherein inhibitors in complexes with target proteins showed stability inside the active site during the 100 ns simulations. Structural changes due to ligand movement inside the cavity were measured in the protein targets, but they remained static due to hydrogen bonds. The results showed acceptable pharmacokinetic properties for resveratrol as compared to penicillin. Thus, we concluded that resveratrol has protective effects against Staphylococcus aureus infections and that it suppresses the quorum-sensing ability of this bacterium by targeting its infectious proteins.

Rosetta's Predictive Ability for Low-Affinity Ligand Binding in Fragment-Based Drug Discovery

Fragment-based drug discovery begins with the identification of small molecules with a molecular weight of usually less than 250 Da which weakly bind to the protein of interest. This technique is challenging for computational docking methods as binding is determined by only a few specific interactions. Inaccuracies in the energy function or slight deviations in the docking pose can lead to the prediction of incorrect binding or difficulties in ranking fragments in in silico screening. Here, we test RosettaLigand by docking a series of fragments to a cysteine-depleted variant of the TIM-barrel protein, HisF (UniProtKB Q9X0C6). We compare the computational results with experimental NMR spectroscopy screens. NMR spectroscopy gives details on binding affinities of individual ligands, which allows assessment of the ligand-ranking ability using RosettaLigand and also provides feedback on the location of the binding pocket, which serves as a reliable test of RosettaLigand's ability to identify plausible binding poses. From a library screen of 3456 fragments, we identified a set of 31 ligands with intrinsic affinities to HisF with dissociation constants as low as 400 μM. The same library of fragments was blindly screened in silico. RosettaLigand was able to rank binders before non-binders with an area under the curve of the receiver operating characteristics of 0.74. The docking poses observed for binders agreed with the binding pocket identified by NMR chemical shift perturbations for all fragments. Taken together, these results provide a baseline performance of RosettaLigand in a fragment-based drug discovery setting.

Protein-Ligand Docking in the Machine-Learning Era

Molecular docking plays a significant role in early-stage drug discovery, from structure-based virtual screening (VS) to hit-to-lead optimization, and its capability and predictive power is critically dependent on the protein-ligand scoring function. In this review, we give a broad overview of recent scoring function development, as well as the docking-based applications in drug discovery. We outline the strategies and resources available for structure-based VS and discuss the assessment and development of classical and machine learning protein-ligand scoring functions. In particular, we highlight the recent progress of machine learning scoring function ranging from descriptor-based models to deep learning approaches. We also discuss the general workflow and docking protocols of structure-based VS, such as structure preparation, binding site detection, docking strategies, and post-docking filter/re-scoring, as well as a case study on the large-scale docking-based VS test on the LIT-PCBA data set.

Introduction to the BioChemical Library (BCL): An Application-Based Open-Source Toolkit for Integrated Cheminformatics and Machine Learning in Computer-Aided Drug Discovery

The BioChemical Library (BCL) cheminformatics toolkit is an application-based academic open-source software package designed to integrate traditional small molecule cheminformatics tools with machine learning-based quantitative structure-activity/property relationship (QSAR/QSPR) modeling. In this pedagogical article we provide a detailed introduction to core BCL cheminformatics functionality, showing how traditional tasks (e.g., computing chemical properties, estimating druglikeness) can be readily combined with machine learning. In addition, we have included multiple examples covering areas of advanced use, such as reaction-based library design. We anticipate that this manuscript will be a valuable resource for researchers in computer-aided drug discovery looking to integrate modular cheminformatics and machine learning tools into their pipelines.

From Drug Molecules to Thermoset Shape Memory Polymers: A Machine Learning Approach

Ultraviolet (UV)-curable thermoset shape memory polymers (TSMPs) with high recovery stress but mild glass transition temperature (T_g) are highly desired for 3D/4D printing lightweight load-bearing structures and devices. However, a bottleneck is that high recovery stress usually means high T_g. For a few TSMPs with high recovery stress, their T_g values are close to the decomposition temperature, and thus, the shape memory effect cannot be triggered safely and effectively. While machine learning (ML) has served as a useful tool to discover new materials and drugs, the grand challenge of using ML to discover new TSMPs persists in the very limited data available. Here, we report an enhanced ML approach by combining the transfer learning-variational autoencoder with a weighted-vector combination method. By learning a large data set with drug molecules in a pretraining process, we were able to effectively map the TSMPs to a hidden space that is much closer to a Gaussian distribution. Through this approach, we created a large compositional space and were able to discover five new types of UV-curable TSMPs with desired properties, one of which was validated by the experiments. Our contribution includes (1) representing the features of TSMPs by drug molecules to overcome the barrier of a limited training data set and (2) developing a ML framework that is able to overcome the barrier of mapping the molar ratio information. It is shown that this approach can effectively learn TSMP features by utilizing the relatedness between the data-scarce (and biased) TSMP target and data-abundant drug source, and the result is much more accurate and more robust than the benchmark set by the support vector machine method using direct label encoding and Morgan encoding. Therefore, it is believed that this framework is a state-of-the-art study in the TSMP field. This study opens new opportunities for discovering not only new TSMPs but also other thermoset polymers.

Identification of Novel HBV/HDV Entry Inhibitors by Pharmacophore- and QSAR-Guided Virtual Screening

The hepatic bile acid transporter Na⁺/taurocholate co-transporting polypeptide (NTCP) was identified in 2012 as the high-affinity hepatic receptor for the hepatitis B and D viruses (HBV/HDV). Since then, this carrier has emerged as promising drug target for HBV/HDV virus entry inhibitors, but the synthetic peptide Hepcludex^® of high molecular weight is the only approved HDV entry inhibitor so far. The present study aimed to identify small molecules as novel NTCP inhibitors with anti-viral activity. A ligand-based bioinformatic approach was used to generate and validate appropriate pharmacophore and QSAR (quantitative structure–activity relationship) models. Half-maximal inhibitory concentrations (IC₅₀) for binding inhibition of the HBV/HDV-derived preS1 peptide (as surrogate parameter for virus binding to NTCP) were determined in NTCP-expressing HEK293 cells for 150 compounds of different chemical classes. IC₅₀ values ranged from 2 µM up to >1000 µM. The generated pharmacophore and QSAR models were used for virtual screening of drug-like chemicals from the ZINC¹⁵ database (~11 million compounds). The 20 best-performing compounds were then experimentally tested for preS1-peptide binding inhibition in NTCP-HEK293 cells. Among them, four compounds were active and revealed experimental IC₅₀ values for preS1-peptide binding inhibition of 9, 19, 20, and 35 µM, which were comparable to the QSAR-based predictions. All these compounds also significantly inhibited in vitro HDV infection of NTCP-HepG2 cells, without showing any cytotoxicity. The best-performing compound in all assays was ZINC000253533654. In conclusion, the present study demonstrates that virtual compound screening based on NTCP-specific pharmacophore and QSAR models can predict novel active hit compounds for the development of HBV/HDV entry inhibitors.

Hepatitis D Virus Entry Inhibitors Based on Repurposing Intestinal Bile Acid Reabsorption Inhibitors

Identification of Na⁺/taurocholate co-transporting polypeptide (NTCP) as high-affinity hepatic entry receptor for the Hepatitis B and D viruses (HBV/HDV) opened the field for target-based development of cell-entry inhibitors. However, most of the HBV/HDV entry inhibitors identified so far also interfere with the physiological bile acid transporter function of NTCP. The present study aimed to identify more virus-selective inhibitors of NTCP by screening of 87 propanolamine derivatives from the former development of intestinal bile acid reabsorption inhibitors (BARIs), which interact with the NTCP-homologous intestinal apical sodium-dependent bile acid transporter (ASBT). In NTCP-HEK293 cells, the ability of these compounds to block the HBV/HDV-derived preS1-peptide binding to NTCP (virus receptor function) as well as the taurocholic acid transport via NTCP (bile acid transporter function) were analyzed in parallel. Hits were subsequently validated by performing in vitro HDV infection experiments in NTCP-HepG2 cells. The most potent compounds S985852, A000295231, and S973509 showed in vitro anti-HDV activities with IC₅₀ values of 15, 40, and 70 µM, respectively, while the taurocholic acid uptake inhibition occurred at much higher IC₅₀ values of 24, 780, and 490 µM, respectively. In conclusion, repurposing of compounds from the BARI class as novel HBV/HDV entry inhibitors seems possible and even enables certain virus selectivity based on structure-activity relationships.

General Purpose Structure-Based Drug Discovery Neural Network Score Functions with Human-Interpretable Pharmacophore Maps

The BioChemical Library (BCL) is an academic open-source cheminformatics toolkit comprising ligand-based virtual high-throughput screening (vHTS) tools such as quantitative structure-activity/property relationship (QSAR/QSPR) modeling, small molecule flexible alignment, small molecule conformer generation, and more. Here, we expand the capabilities of the BCL to include structure-based virtual screening. We introduce two new score functions, BCL-AffinityNet and BCL-DockANNScore, based on novel distance-dependent signed protein-ligand atomic property correlations. Both metrics are conventional feed-forward dropout neural networks trained on the new descriptors. We demonstrate that BCL-AffinityNet is one of the top performing score functions on the comparative assessment of score functions 2016 affinity prediction and affinity ranking tasks. We also demonstrate that BCL-AffinityNet performs well on the CSAR-NRC HiQ I and II test sets. Furthermore, we demonstrate that BCL-DockANNScore is competitive with multiple state-of-the-art methods on the docking power and screening power tasks. Finally, we show how our models can be decomposed into human-interpretable pharmacophore maps to aid in hit/lead optimization. Altogether, our results expand the utility of the BCL for structure-based scoring to aid small molecule discovery and design. BCL-AffinityNet, BCL-DockANNScore, and the pharmacophore mapping application, as well as the remainder of the BCL cheminformatics toolkit, are freely available with an academic license at the BCL Commons site hosted on http://meilerlab.org/.

BCL::Conf: Improved Open-Source Knowledge-Based Conformation Sampling Using the Crystallography Open Database

We previously described BCL::Conf, a knowledge-based conformation sampling algorithm utilizing a small molecule fragment rotamer library derived from the Cambridge Structural Database (CSD, license required), as a component of the BioChemical Library (BCL) cheminformatics toolkit. This paper describes substantial improvements made to the BCL::Conf algorithm and a transition to a rotamer library derived from molecules in the Crystallography Open Database (COD, no license required). We demonstrate the performance of the new BCL::Conf on native conformer recovery in the Platinum dataset of high-quality protein-ligand complexes. This set of 2859 structures has previously been used to assess the performance of over a dozen conformer generation algorithms, including the Conformator, Balloon, RDKit DG, ETKDG, Confab, Frog2, MultiConf-DOCK, CSD conformer generator, ConfGenX-OPSL3 force field, Omega, excalc, iCon, and MOE. These benchmarks suggest that the CSD conformer generator is at the state of the art of reported conformer generators. Our results indicate that the improved BCL::Conf significantly outperforms the CSD conformer generation algorithm at binding conformer recovery across a range of ensemble sizes and with similarly fast rates of conformer generation. BCL::Conf is now distributed with the COD-derived rotamer library and is free for academic use. The BCL can be downloaded at http://meilerlab.org/bclcommons for Windows, Linux, or Apple operating systems. BCL::Conf can now also be accessed via webserver at http://meilerlab.org/bclconf.

Ranking Molecules with Vanishing Kernels and a Single Parameter: Active Applicability Domain Included

In ligand-based virtual screening, high-throughput screening (HTS) data sets can be exploited to train classification models. Such models can be used to prioritize yet untested molecules, from the most likely active (against a protein target of interest) to the least likely active. In this study, a single-parameter ranking method with an Applicability Domain (AD) is proposed. In effect, Kernel Density Estimates (KDE) are revisited to improve their computational efficiency and incorporate an AD. Two modifications are proposed: (i) using vanishing kernels (i.e., kernel functions with a finite support) and (ii) using the Tanimoto distance between molecular fingerprints as a radial basis function. This construction is termed "Vanishing Ranking Kernels" (VRK). Using VRK on 21 HTS assays, it is shown that VRK can compete in performance with a graph convolutional deep neural network. VRK are conceptually simple and fast to train. During training, they require optimizing a single parameter. A trained VRK model usually defines an active AD. Exploiting this AD can significantly increase the screening frequency of a VRK model. Software: https://github.com/UnixJunkie/rankers. Data sets: https://zenodo.org/record/1320776 and https://zenodo.org/record/3540423.

Benchmarking Data Sets from PubChem BioAssay Data: Current Scenario and Room for Improvement

Developing realistic data sets for evaluating virtual screening methods is a task that has been tackled by the cheminformatics community for many years. Numerous artificially constructed data collections were developed, such as DUD, DUD-E, or DEKOIS. However, they all suffer from multiple drawbacks, one of which is the absence of experimental results confirming the impotence of presumably inactive molecules, leading to possible false negatives in the ligand sets. In light of this problem, the PubChem BioAssay database, an open-access repository providing the bioactivity information of compounds that were already tested on a biological target, is now a recommended source for data set construction. Nevertheless, there exist several issues with the use of such data that need to be properly addressed. In this article, an overview of benchmarking data collections built upon experimental PubChem BioAssay input is provided, along with a thorough discussion of noteworthy issues that one must consider during the design of new ligand sets from this database. The points raised in this review are expected to guide future developments in this regard, in hopes of offering better evaluation tools for novel in silico screening procedures.

Combining Docking Pose Rank and Structure with Deep Learning Improves Protein-Ligand Binding Mode Prediction over a Baseline Docking Approach

We present a simple, modular graph-based convolutional neural network that takes structural information from protein-ligand complexes as input to generate models for activity and binding mode prediction. Complex structures are generated by a standard docking procedure and fed into a dual-graph architecture that includes separate subnetworks for the ligand bonded topology and the ligand-protein contact map. Recent work has indicated that data set bias drives many past promising results derived from combining deep learning and docking. Our dual-graph network allows contributions from ligand identity that give rise to such biases to be distinguished from effects of protein-ligand interactions on classification. We show that our neural network is capable of learning from protein structural information when, as in the case of binding mode prediction, an unbiased data set is constructed. We next develop a deep learning model for binding mode prediction that uses docking ranking as input in combination with docking structures. This strategy mirrors past consensus models and outperforms a baseline docking program (AutoDock Vina) in a variety of tests, including on cross-docking data sets that mimic real-world docking use cases. Furthermore, the magnitudes of network predictions serve as reliable measures of model confidence.

Cinnamic Derivatives as Antitubercular Agents: Characterization by Quantitative Structure-Activity Relationship Studies

Tuberculosis, caused by Mycobacterium tuberculosis (Mtb), remains one of the top ten causes of death worldwide and the main cause of mortality from a single infectious agent. The upsurge of multi- and extensively-drug resistant tuberculosis cases calls for an urgent need to develop new and more effective antitubercular drugs. As the cinnamoyl scaffold is a privileged and important pharmacophore in medicinal chemistry, some studies were conducted to find novel cinnamic acid derivatives (CAD) potentially active against tuberculosis. In this context, we have engaged in the setting up of a quantitative structure–activity relationships (QSAR) strategy to: (i) derive through multiple linear regression analysis a statistically significant model to describe the antitubercular activity of CAD towards wild-type Mtb; and (ii) identify the most relevant properties with an impact on the antitubercular behavior of those derivatives. The best-found model involved only geometrical and electronic CAD related properties and was successfully challenged through strict internal and external validation procedures. The physicochemical information encoded by the identified descriptors can be used to propose specific structural modifications to design better CAD antitubercular compounds.

Identification of Novel Allosteric Modulators of Metabotropic Glutamate Receptor Subtype 5 Acting at Site Distinct from 2-Methyl-6-(phenylethynyl)-pyridine Binding

As part of the G-protein coupled receptor (GPCR) family, metabotropic glutamate (mGlu) receptors play an important role as drug targets of cognitive diseases. Selective allosteric modulators of mGlu subtype 5 (mGlu₅) have the potential to alleviate symptoms of numerous central nervous system disorders such as schizophrenia in a more targeted fashion. Multiple mGlu₅ positive allosteric modulators (PAMs), such as 1-(3-fluorophenyl)-N-((3-fluorophenyl)-methylideneamino)-methanimine (DFB), 3-cyano-N-(1,3-diphenyl-1H-pyrazol-5-yl)-benzamide (CDPPB), and 4-nitro-N-(1,3-diphenyl-1H-pyrazol-5-yl)-benzamide (VU-29), exert their actions by binding to a defined allosteric site on mGlu₅ located in the seven-transmembrane domain (7TM) and shared by mGlu₅ negative allosteric modulator (NAM) 2-methyl-6-(phenylethynyl)-pyridine (MPEP). Actions of the PAM N-{4-chloro-2-[(1,3-dioxo-1,3-dihydro-2H-isoindol-2-yl)methyl]phenyl}-2-hydroxybenzamide (CPPHA) are mediated by a distinct allosteric site in the 7TM domain different from the MPEP binding site. Experimental evidence confirms these findings through mutagenesis experiments involving residues F585 (TM1) and A809 (TM7). In an effort to investigate mGlu₅ PAM selectivity for this alternative allosteric site distinct from MPEP binding, we employed in silico quantitative structure-activity relationship (QSAR) modeling. Subsequent ligand-based virtual screening prioritized a set of 63 candidate compounds predicted from a library of over 4 million commercially available compounds to bind exclusively to this novel site. Experimental validation verified the biological activity for seven of 63 selected candidates. Further, medicinal chemistry optimizations based on these molecules revealed compound VU6003586 with an experimentally validated potency of 174 nM. Radioligand binding experiments showed only partial inhibition at very high concentrations, most likely indicative of binding at a non-MPEP site. Selective positive allosteric modulators for mGlu₅ have the potential for tremendous impact concerning devastating neurological disorders such as schizophrenia and Huntington's disease. These identified and validated novel selective compounds can serve as starting points for more specifically tailored lead and probe molecules and thus help the development of potential therapeutic agents with reduced adverse effects.

In Silico Toxicology Data Resources to Support Read-Across and (Q)SAR

A plethora of databases exist online that can assist in in silico chemical or drug safety assessment. However, a systematic review and grouping of databases, based on purpose and information content, consolidated in a single source, has been lacking. To resolve this issue, this review provides a comprehensive listing of the key in silico data resources relevant to: chemical identity and properties, drug action, toxicology (including nano-material toxicity), exposure, omics, pathways, Absorption, Distribution, Metabolism and Elimination (ADME) properties, clinical trials, pharmacovigilance, patents-related databases, biological (genes, enzymes, proteins, other macromolecules etc.) databases, protein-protein interactions (PPIs), environmental exposure related, and finally databases relating to animal alternatives in support of 3Rs policies. More than nine hundred databases were identified and reviewed against criteria relating to accessibility, data coverage, interoperability or application programming interface (API), appropriate identifiers, types of in vitro, in vivo,-clinical or other data recorded and suitability for modelling, read-across, or similarity searching. This review also specifically addresses the need for solutions for mapping and integration of databases into a common platform for better translatability of preclinical data to clinical data.

BCL::Mol2D-a robust atom environment descriptor for QSAR modeling and lead optimization

Comparing fragment based molecular fingerprints of drug-like molecules is one of the most robust and frequently used approaches in computer-assisted drug discovery. Molprint2D, a popular atom environment (AE) descriptor, yielded the best enrichment of active compounds across a diverse set of targets in a recent large-scale study. We present here BCL::Mol2D descriptors that outperformed Molprint2D on nine PubChem datasets spanning a wide range of protein classes. Because BCL::Mol2D records the number of AEs from a universal AE library, a novel aspect of BCL::Mol2D over the Molprint2D is its reversibility. This property enables decomposition of prediction from machine learning models to particular molecular substructures. Artificial neural networks with dropout, when trained on BCL::Mol2D descriptors outperform those trained on Molprint2D descriptors by up to 26% in logAUC metric. When combined with the Reduced Short Range descriptor set, our previously published set of descriptors optimized for QSARs, BCL::Mol2D yields a modest improvement. Finally, we demonstrate how the reversibility of BCL::Mol2D enables visualization of a 'pharmacophore map' that could guide lead optimization for serine/threonine kinase 33 inhibitors.

BCL::MolAlign: Three-Dimensional Small Molecule Alignment for Pharmacophore Mapping

Small molecule flexible alignment is a critical component of both ligand- and structure-based methods in computer-aided drug discovery. Despite its importance, the availability of high-quality flexible alignment software packages is limited. Here, we present BCL::MolAlign, a freely available property-based molecular alignment program. BCL::MolAlign accommodates ligand flexibility through a combination of pregenerated conformers and on-the-fly bond rotation. BCL::MolAlign converges on alignment poses by sampling the relative orientations of mutually matching atom pairs between molecules through Monte Carlo Metropolis sampling. Across six diverse ligand data sets, BCL::MolAlign flexible alignment outperforms MOE, ROCS, and FLEXS in recovering native ligand binding poses. Moreover, the BCL::MolAlign alignment score is more predictive of ligand activity than maximum common substructure similarity across 10 data sets. Finally, on a recently published benchmark set of 20 high quality congeneric ligand-protein complexes, BCL::MolAlign is able to recover a larger fraction of native binding poses than maximum common substructure-based alignment and RosettaLigand. BCL::MolAlign can be obtained as part of the Biology and Chemistry Library (BCL) software package freely with an academic license or can be accessed via Web server at http://meilerlab.org/index.php/servers/molalign .

A Distance-Based Boolean Applicability Domain for Classification of High Throughput Screening Data

In Quantitative Structure-Activity Relationship (QSAR) modeling, one must come up with an activity model but also with an applicability domain for that model. Some existing methods to create an applicability domain are complex, hard to implement, and/or difficult to interpret. Also, they often require the user to select a threshold value, or they embed an empirical constant. In this work, we propose a trivial to interpret and fully automatic Distance-Based Boolean Applicability Domain (DBBAD) algorithm for category QSAR. In retrospective experiments on High Throughput Screening data sets, this applicability domain improves the classification performance and early retrieval of support vector machine and random forest based classifiers, while improving the scaffold diversity among top-ranked active molecules.

Advances with support vector machines for novel drug discovery

INTRODUCTION

Novel drug discovery remains an enormous challenge, with various computer-aided drug design (CADD) approaches having been widely employed for this purpose. CADD, specifically the commonly used support vector machines (SVMs), can employ machine learning techniques. SVMs and their variations offer numerous drug discovery applications, which range from the classification of substances (as active or inactive) to the construction of regression models and the ranking/virtual screening of databased compounds. Areas covered: Herein, the authors consider some of the applications of SVMs in medicinal chemistry, illustrating their main advantages and disadvantages, as well as trends in their utilization, via the available published literature. The aim of this review is to provide an up-to-date review of the recent applications of SVMs in drug discovery as described by the literature, thereby highlighting their strengths, weaknesses, and future challenges. Expert opinion: Techniques based on SVMs are considered as powerful approaches in early drug discovery. The ability of SVMs to classify active or inactive compounds has enabled the prioritization of substances for virtual screening. Indeed, one of the main advantages of SVMs is related to their potential in the analysis of nonlinear problems. However, despite successes in employing SVMs, the challenges of improving accuracy remain.

QSAR-Based Virtual Screening: Advances and Applications in Drug Discovery

Virtual screening (VS) has emerged in drug discovery as a powerful computational approach to screen large libraries of small molecules for new hits with desired properties that can then be tested experimentally. Similar to other computational approaches, VS intention is not to replace in vitro or in vivo assays, but to speed up the discovery process, to reduce the number of candidates to be tested experimentally, and to rationalize their choice. Moreover, VS has become very popular in pharmaceutical companies and academic organizations due to its time-, cost-, resources-, and labor-saving. Among the VS approaches, quantitative structure–activity relationship (QSAR) analysis is the most powerful method due to its high and fast throughput and good hit rate. As the first preliminary step of a QSAR model development, relevant chemogenomics data are collected from databases and the literature. Then, chemical descriptors are calculated on different levels of representation of molecular structure, ranging from 1D to nD, and then correlated with the biological property using machine learning techniques. Once developed and validated, QSAR models are applied to predict the biological property of novel compounds. Although the experimental testing of computational hits is not an inherent part of QSAR methodology, it is highly desired and should be performed as an ultimate validation of developed models. In this mini-review, we summarize and critically analyze the recent trends of QSAR-based VS in drug discovery and demonstrate successful applications in identifying perspective compounds with desired properties. Moreover, we provide some recommendations about the best practices for QSAR-based VS along with the future perspectives of this approach.

Predicting the Functional Impact of KCNQ1 Variants of Unknown Significance

BACKGROUND

An emerging standard-of-care for long-QT syndrome uses clinical genetic testing to identify genetic variants of the KCNQ1 potassium channel. However, interpreting results from genetic testing is confounded by the presence of variants of unknown significance for which there is inadequate evidence of pathogenicity.

METHODS AND RESULTS

In this study, we curated from the literature a high-quality set of 107 functionally characterized KCNQ1 variants. Based on this data set, we completed a detailed quantitative analysis on the sequence conservation patterns of subdomains of KCNQ1 and the distribution of pathogenic variants therein. We found that conserved subdomains generally are critical for channel function and are enriched with dysfunctional variants. Using this experimentally validated data set, we trained a neural network, designated Q1VarPred, specifically for predicting the functional impact of KCNQ1 variants of unknown significance. The estimated predictive performance of Q1VarPred in terms of Matthew's correlation coefficient and area under the receiver operating characteristic curve were 0.581 and 0.884, respectively, superior to the performance of 8 previous methods tested in parallel. Q1VarPred is publicly available as a web server at http://meilerlab.org/q1varpred.

CONCLUSIONS

Although a plethora of tools are available for making pathogenicity predictions over a genome-wide scale, previous tools fail to perform in a robust manner when applied to KCNQ1. The contrasting and favorable results for Q1VarPred suggest a promising approach, where a machine-learning algorithm is tailored to a specific protein target and trained with a functionally validated data set to calibrate informatics tools.

Selecting Feature Subsets Based on SVM-RFE and the Overlapping Ratio with Applications in Bioinformatics

Feature selection is an important topic in bioinformatics. Defining informative features from complex high dimensional biological data is critical in disease study, drug development, etc. Support vector machine-recursive feature elimination (SVM-RFE) is an efficient feature selection technique that has shown its power in many applications. It ranks the features according to the recursive feature deletion sequence based on SVM. In this study, we propose a method, SVM-RFE-OA, which combines the classification accuracy rate and the average overlapping ratio of the samples to determine the number of features to be selected from the feature rank of SVM-RFE. Meanwhile, to measure the feature weights more accurately, we propose a modified SVM-RFE-OA (M-SVM-RFE-OA) algorithm that temporally screens out the samples lying in a heavy overlapping area in each iteration. The experiments on the eight public biological datasets show that the discriminative ability of the feature subset could be measured more accurately by combining the classification accuracy rate with the average overlapping degree of the samples compared with using the classification accuracy rate alone, and shielding the samples in the overlapping area made the calculation of the feature weights more stable and accurate. The methods proposed in this study can also be used with other RFE techniques to define potential biomarkers from big biological data.

Consensus queries in ligand-based virtual screening experiments

Background

In ligand-based virtual screening experiments, a known active ligand is used in similarity searches to find putative active compounds for the same protein target. When there are several known active molecules, screening using all of them is more powerful than screening using a single ligand. A consensus query can be created by either screening serially with different ligands before merging the obtained similarity scores, or by combining the molecular descriptors (i.e. chemical fingerprints) of those ligands.

Results

We report on the discriminative power and speed of several consensus methods, on two datasets only made of experimentally verified molecules. The two datasets contain a total of 19 protein targets, 3776 known active and ~ 2 × 10⁶ inactive molecules. Three chemical fingerprints are investigated: MACCS 166 bits, ECFP4 2048 bits and an unfolded version of MOLPRINT2D. Four different consensus policies and five consensus sizes were benchmarked.

Conclusions

The best consensus method is to rank candidate molecules using the maximum score obtained by each candidate molecule versus all known actives. When the number of actives used is small, the same screening performance can be approached by a consensus fingerprint. However, if the computational exploration of the chemical space is limited by speed (i.e. throughput), a consensus fingerprint allows to outperform this consensus of scores.

Quantitative Structure-Activity Relationship Modeling of Kinase Selectivity Profiles

The discovery of selective inhibitors of biological target proteins is the primary goal of many drug discovery campaigns. However, this goal has proven elusive, especially for inhibitors targeting the well-conserved orthosteric adenosine triphosphate (ATP) binding pocket of kinase enzymes. The human kinome is large and it is rather difficult to profile early lead compounds against around 500 targets to gain an upfront knowledge on selectivity. Further, selectivity can change drastically during derivatization of an initial lead compound. Here, we have introduced a computational model to support the profiling of compounds early in the drug discovery pipeline. On the basis of the extensive profiled activity of 70 kinase inhibitors against 379 kinases, including 81 tyrosine kinases, we developed a quantitative structure-activity relation (QSAR) model using artificial neural networks, to predict the activity of these kinase inhibitors against the panel of 379 kinases. The model's performance in predicting activity ranges from 0.6 to 0.8 depending on the kinase, from the area under the curve (AUC) of the receiver operating characteristics (ROC). The profiler is available online at http://www.meilerlab.org/index.php/servers/show?s_id=23.

High-Throughput Screening Assay Datasets from the PubChem Database

Availability of high-throughput screening (HTS) data in the public domain offers great potential to foster development of ligand-based computer-aided drug discovery (LB-CADD) methods crucial for drug discovery efforts in academia and industry. LB-CADD method development depends on high-quality HTS assay data, i.e., datasets that contain both active and inactive compounds. These active compounds are hits from primary screens that have been tested in concentration-response experiments and where the target-specificity of the hits has been validated through suitable secondary screening experiments. Publicly available HTS repositories such as PubChem often provide such data in a convoluted way: compounds that are classified as inactive need to be extracted from the primary screening record. However, compounds classified as active in the primary screening record are not suitable as a set of active compounds for LB-CADD experiments due to high false-positive rate. A suitable set of actives can be derived by carefully analysing results in often up to five or more assays that are used to confirm and classify the activity of compounds. These assays, in part, build on each other. However, often not all hit compounds from the previous screen have been tested. Sometimes a compound can be classified as 'active', though its meaning is 'inactive' on the target of interest as it is 'active' on a different target protein. Here, a curation process of hierarchically related confirmatory screens is illustrated based on two specifically chosen protein use-cases. The subsequent re-upload procedure into PubChem is described for the findings of those two scenarios. Further, we provide nine publicly accessible high quality datasets for future LB-CADD method development that provide a common baseline for comparison of future methods to the scientific community. We also provide a protocol researchers can follow to upload additional datasets for benchmarking.

An Introduction to the Basic Concepts in QSAR-Aided Drug Design

The need for the development of new drugs to combat existing and newly identified conditions is unavoidable. One of the important tools used in the advanced drug development pipeline is computer-aided drug design. Traditionally, to find a drug many ligands were synthesized and evaluated for their effectiveness using suitable bioassays and if all other drug-likeness features were met, the candidate(s) would possibly reach the market. Although this approach is still in use in advanced format, computational methods are an indispensable component of modern drug development projects. One of the methods used from very early days of rationalizing the drug design approaches is Quantitative Structure-Activity Relationship (QSAR). This chapter overviews QSAR modeling steps by introducing molecular descriptors, mathematical model development for relating biological activities to molecular structures, and model validation. At the end, several successful cases where QSAR studies were used extensively are presented.

Ligand-based virtual screen for the discovery of novel M5 inhibitor chemotypes

This Letter describes a ligand-based virtual screening campaign utilizing SAR data around the M5 NAMs, ML375 and VU6000181. Both QSAR and shape scores were employed to virtually screen a 98,000-member compound library. Neither approach alone proved productive, but a consensus score of the two models identified a novel scaffold which proved to be a modestly selective, but weak inhibitor (VU0549108) of the M5 mAChR (M5 IC50=6.2μM, M1-4 IC50s>10μM) based on an unusual 8-((1,3,5-trimethyl-1H-pyrazol-4-yl)sulfonyl)-1-oxa-4-thia-8-azaspiro[4,5]decane scaffold. [(3)H]-NMS binding studies showed that VU0549108 interacts with the orthosteric site (Ki of 2.7μM), but it is not clear if this is negative cooperativity or orthosteric binding. Interestingly, analogs synthesized around VU0549108 proved weak, and SAR was very steep. However, this campaign validated the approach and warranted further expansion to identify additional novel chemotypes.

Getting the most out of PubChem for virtual screening

INTRODUCTION

With the emergence of the 'big data' era, the biomedical research community has great interest in exploiting publicly available chemical information for drug discovery. PubChem is an example of public databases that provide a large amount of chemical information free of charge.

AREAS COVERED

This article provides an overview of how PubChem's data, tools, and services can be used for virtual screening and reviews recent publications that discuss important aspects of exploiting PubChem for drug discovery.

EXPERT OPINION

PubChem offers comprehensive chemical information useful for drug discovery. It also provides multiple programmatic access routes, which are essential to build automated virtual screening pipelines that exploit PubChem data. In addition, PubChemRDF allows users to download PubChem data and load them into a local computing facility, facilitating data integration between PubChem and other resources. PubChem resources have been used in many studies for developing bioactivity and toxicity prediction models, discovering polypharmacologic (multi-target) ligands, and identifying new macromolecule targets of compounds (for drug-repurposing or off-target side effect prediction). These studies demonstrate the usefulness of PubChem as a key resource for computer-aided drug discovery and related area.

Improving quantitative structure-activity relationship models using Artificial Neural Networks trained with dropout

Dropout is an Artificial Neural Network (ANN) training technique that has been shown to improve ANN performance across canonical machine learning (ML) datasets. Quantitative Structure Activity Relationship (QSAR) datasets used to relate chemical structure to biological activity in Ligand-Based Computer-Aided Drug Discovery pose unique challenges for ML techniques, such as heavily biased dataset composition, and relatively large number of descriptors relative to the number of actives. To test the hypothesis that dropout also improves QSAR ANNs, we conduct a benchmark on nine large QSAR datasets. Use of dropout improved both enrichment false positive rate and log-scaled area under the receiver-operating characteristic curve (logAUC) by 22-46 % over conventional ANN implementations. Optimal dropout rates are found to be a function of the signal-to-noise ratio of the descriptor set, and relatively independent of the dataset. Dropout ANNs with 2D and 3D autocorrelation descriptors outperform conventional ANNs as well as optimized fingerprint similarity search methods.

Autocorrelation descriptor improvements for QSAR: 2DA_Sign and 3DA_Sign

Quantitative structure-activity relationship (QSAR) is a branch of computer aided drug discovery that relates chemical structures to biological activity. Two well established and related QSAR descriptors are two- and three-dimensional autocorrelation (2DA and 3DA). These descriptors encode the relative position of atoms or atom properties by calculating the separation between atom pairs in terms of number of bonds (2DA) or Euclidean distance (3DA). The sums of all values computed for a given small molecule are collected in a histogram. Atom properties can be added with a coefficient that is the product of atom properties for each pair. This procedure can lead to information loss when signed atom properties are considered such as partial charge. For example, the product of two positive charges is indistinguishable from the product of two equivalent negative charges. In this paper, we present variations of 2DA and 3DA called 2DA_Sign and 3DA_Sign that avoid information loss by splitting unique sign pairs into individual histograms. We evaluate these variations with models trained on nine datasets spanning a range of drug target classes. Both 2DA_Sign and 3DA_Sign significantly increase model performance across all datasets when compared with traditional 2DA and 3DA. Lastly, we find that limiting 3DA_Sign to maximum atom pair distances of 6 Å instead of 12 Å further increases model performance, suggesting that conformational flexibility may hinder performance with longer 3DA descriptors. Consistent with this finding, limiting the number of bonds in 2DA_Sign from 11 to 5 fails to improve performance.

Combinatorial Consensus Scoring for Ligand-Based Virtual Fragment Screening: A Comparative Case Study for Serotonin 5-HT(3)A, Histamine H(1), and Histamine H(4) Receptors

In the current study we have evaluated the applicability of ligand-based virtual screening (LBVS) methods for the identification of small fragment-like biologically active molecules using different similarity descriptors and different consensus scoring approaches. For this purpose, we have evaluated the performance of 14 chemical similarity descriptors in retrospective virtual screening studies to discriminate fragment-like ligands of three membrane-bound receptors from fragments that are experimentally determined to have no affinity for these proteins (true inactives). We used a complete fragment affinity data set of experimentally determined ligands and inactives for two G protein-coupled receptors (GPCRs), the histamine H1 receptor (H1R) and the histamine H4 receptor (H4R), and one ligand-gated ion channel (LGIC), the serotonin receptor (5-HT3AR), to validate our retrospective virtual screening studies. We have exhaustively tested consensus scoring strategies that combine the results of multiple actives (group fusion) or combine different similarity descriptors (similarity fusion), and for the first time systematically evaluated different combinations of group fusion and similarity fusion approaches. Our studies show that for these three case study protein targets both consensus scoring approaches can increase virtual screening enrichments compared to single chemical similarity search methods. Our cheminformatics analyses recommend to use a combination of both group fusion and similarity fusion for prospective ligand-based virtual fragment screening.

Perspective on computational and structural aspects of kinase discovery from IPK2014

Recent advances in understanding the activity and selectivity of kinase inhibitors and their relationships to protein structure are presented. Conformational selection in kinases is studied from empirical, data-driven and simulation approaches. Ligand binding and its affinity are, in many cases, determined by the predetermined active and inactive conformation of kinases. Binding affinity and selectivity predictions highlight the current state of the art and advances in computational chemistry as it applies to kinase inhibitor discovery. Kinome wide inhibitor profiling and cell panel profiling lead to a better understanding of selectivity and allow for target validation and patient tailoring hypotheses. This article is part of a Special Issue entitled: Inhibitors of Protein Kinases.

Complete genome-wide screening and subtractive genomic approach revealed new virulence factors, potential drug targets against bio-war pathogen Brucella melitensis 16M

Brucella melitensis 16M is a Gram-negative coccobacillus that infects both animals and humans. It causes a disease known as brucellosis, which is characterized by acute febrile illness in humans and causes abortions in livestock. To prevent and control brucellosis, identification of putative drug targets is crucial. The present study aimed to identify drug targets in B. melitensis 16M by using a subtractive genomic approach. We used available database repositories (Database of Essential Genes, Kyoto Encyclopedia of Genes and Genomes Automatic Annotation Server, and Kyoto Encyclopedia of Genes and Genomes) to identify putative genes that are nonhomologous to humans and essential for pathogen B. melitensis 16M. The results revealed that among 3 Mb genome size of pathogen, 53 putative characterized and 13 uncharacterized hypothetical genes were identified; further, from Basic Local Alignment Search Tool protein analysis, one hypothetical protein showed a close resemblance (50%) to Silicibacter pomeroyi DUF1285 family protein (2RE3). A further homology model of the target was constructed using MODELLER 9.12 and optimized through variable target function method by molecular dynamics optimization with simulating annealing. The stereochemical quality of the restrained model was evaluated by PROCHECK, VERIFY-3D, ERRAT, and WHATIF servers. Furthermore, structure-based virtual screening was carried out against the predicted active site of the respective protein using the glycerol structural analogs from the PubChem database. We identified five best inhibitors with strong affinities, stable interactions, and also with reliable drug-like properties. Hence, these leads might be used as the most effective inhibitors of modeled protein. The outcome of the present work of virtual screening of putative gene targets might facilitate design of potential drugs for better treatment against brucellosis.

An Introduction to the Basic Concepts in QSAR-Aided Drug Design

The need for the development of new drugs to combat existing and newly identified conditions is unavoidable. One of the important tools used in the advanced drug development pipeline is computer-aided drug design. Traditionally, to find a drug many ligands were synthesized and evaluated for their effectiveness using suitable bioassays and if all other drug-likeness features were met, the candidate(s) would possibly reach the market. Although this approach is still in use in advanced format, computational methods are an indispensable component of modern drug development projects. One of the methods used from very early days of rationalizing the drug design approaches is Quantitative Structure-Activity Relationship (QSAR). This chapter overviews QSAR modeling steps by introducing molecular descriptors, mathematical model development for relating biological activities to molecular structures, and model validation. At the end, several successful cases where QSAR studies were used extensively are presented.

Computational modeling of membrane proteins

The determination of membrane protein (MP) structures has always trailed that of soluble proteins due to difficulties in their overexpression, reconstitution into membrane mimetics, and subsequent structure determination. The percentage of MP structures in the protein databank (PDB) has been at a constant 1-2% for the last decade. In contrast, over half of all drugs target MPs, only highlighting how little we understand about drug-specific effects in the human body. To reduce this gap, researchers have attempted to predict structural features of MPs even before the first structure was experimentally elucidated. In this review, we present current computational methods to predict MP structure, starting with secondary structure prediction, prediction of trans-membrane spans, and topology. Even though these methods generate reliable predictions, challenges such as predicting kinks or precise beginnings and ends of secondary structure elements are still waiting to be addressed. We describe recent developments in the prediction of 3D structures of both α-helical MPs as well as β-barrels using comparative modeling techniques, de novo methods, and molecular dynamics (MD) simulations. The increase of MP structures has (1) facilitated comparative modeling due to availability of more and better templates, and (2) improved the statistics for knowledge-based scoring functions. Moreover, de novo methods have benefited from the use of correlated mutations as restraints. Finally, we outline current advances that will likely shape the field in the forthcoming decade.

Cellular manganese content is developmentally regulated in human dopaminergic neurons

Manganese (Mn) is both an essential biological cofactor and neurotoxicant. Disruption of Mn biology in the basal ganglia has been implicated in the pathogenesis of neurodegenerative disorders, such as parkinsonism and Huntington's disease. Handling of other essential metals (e.g. iron and zinc) occurs via complex intracellular signaling networks that link metal detection and transport systems. However, beyond several non-selective transporters, little is known about the intracellular processes regulating neuronal Mn homeostasis. We hypothesized that small molecules that modulate intracellular Mn could provide insight into cell-level Mn regulatory mechanisms. We performed a high throughput screen of 40,167 small molecules for modifiers of cellular Mn content in a mouse striatal neuron cell line. Following stringent validation assays and chemical informatics, we obtained a chemical 'toolbox' of 41 small molecules with diverse structure-activity relationships that can alter intracellular Mn levels under biologically relevant Mn exposures. We utilized this toolbox to test for differential regulation of Mn handling in human floor-plate lineage dopaminergic neurons, a lineage especially vulnerable to environmental Mn exposure. We report differential Mn accumulation between developmental stages and stage-specific differences in the Mn-altering activity of individual small molecules. This work demonstrates cell-level regulation of Mn content across neuronal differentiation.

PubChem applications in drug discovery: a bibliometric analysis

A bibliometric analysis of PubChem applications is presented by reviewing 1132 research articles. The massive volume of chemical structure and bioactivity data in PubChem and its online services have been used globally in various fields including chemical biology, medicinal chemistry and informatics research. PubChem supports drug discovery in many aspects such as lead identification and optimization, compound-target profiling, polypharmacology studies and unknown chemical identity elucidation. PubChem has also become a valuable resource for developing secondary databases, informatics tools and web services. The growing PubChem resource with its public availability offers support and great opportunities for the interrogation of pharmacological mechanisms and the genetic basis of diseases, which are vital for drug innovation and repurposing.

Profiling animal toxicants by automatically mining public bioassay data: a big data approach for computational toxicology

In vitro bioassays have been developed and are currently being evaluated as potential alternatives to traditional animal toxicity models. Already, the progress of high throughput screening techniques has resulted in an enormous amount of publicly available bioassay data having been generated for a large collection of compounds. When a compound is tested using a collection of various bioassays, all the testing results can be considered as providing a unique bio-profile for this compound, which records the responses induced when the compound interacts with different cellular systems or biological targets. Profiling compounds of environmental or pharmaceutical interest using useful toxicity bioassay data is a promising method to study complex animal toxicity. In this study, we developed an automatic virtual profiling tool to evaluate potential animal toxicants. First, we automatically acquired all PubChem bioassay data for a set of 4,841 compounds with publicly available rat acute toxicity results. Next, we developed a scoring system to evaluate the relevance between these extracted bioassays and animal acute toxicity. Finally, the top ranked bioassays were selected to profile the compounds of interest. The resulting response profiles proved to be useful to prioritize untested compounds for their animal toxicity potentials and form a potential in vitro toxicity testing panel. The protocol developed in this study could be combined with structure-activity approaches and used to explore additional publicly available bioassay datasets for modeling a broader range of animal toxicities.

An unbiased method to build benchmarking sets for ligand-based virtual screening and its application to GPCRs

Benchmarking data sets have become common in recent years for the purpose of virtual screening, though the main focus had been placed on the structure-based virtual screening (SBVS) approaches. Due to the lack of crystal structures, there is great need for unbiased benchmarking sets to evaluate various ligand-based virtual screening (LBVS) methods for important drug targets such as G protein-coupled receptors (GPCRs). To date these ready-to-apply data sets for LBVS are fairly limited, and the direct usage of benchmarking sets designed for SBVS could bring the biases to the evaluation of LBVS. Herein, we propose an unbiased method to build benchmarking sets for LBVS and validate it on a multitude of GPCRs targets. To be more specific, our methods can (1) ensure chemical diversity of ligands, (2) maintain the physicochemical similarity between ligands and decoys, (3) make the decoys dissimilar in chemical topology to all ligands to avoid false negatives, and (4) maximize spatial random distribution of ligands and decoys. We evaluated the quality of our Unbiased Ligand Set (ULS) and Unbiased Decoy Set (UDS) using three common LBVS approaches, with Leave-One-Out (LOO) Cross-Validation (CV) and a metric of average AUC of the ROC curves. Our method has greatly reduced the “artificial enrichment” and “analogue bias” of a published GPCRs benchmarking set, i.e., GPCR Ligand Library (GLL)/GPCR Decoy Database (GDD). In addition, we addressed an important issue about the ratio of decoys per ligand and found that for a range of 30 to 100 it does not affect the quality of the benchmarking set, so we kept the original ratio of 39 from the GLL/GDD.

Focused chemical libraries--design and enrichment: an example of protein-protein interaction chemical space

One of the many obstacles in the development of new drugs lies in the limited number of therapeutic targets and in the quality of screening collections of compounds. In this review, we present general strategies for building target-focused chemical libraries with a particular emphasis on protein-protein interactions (PPIs). We describe the chemical spaces spanned by nine commercially available PPI-focused libraries and compare them to our 2P2I3D academic library, dedicated to orthosteric PPI modulators. We show that although PPI-focused libraries have been designed using different strategies, they share common subspaces. PPI inhibitors are larger and more hydrophobic than standard drugs; however, an effort has been made to improve the drug-likeness of focused chemical libraries dedicated to this challenging class of targets.

PubChem BioAssay: 2014 update

PubChem’s BioAssay database (http://pubchem.ncbi.nlm.nih.gov) is a public repository for archiving biological tests of small molecules generated through high-throughput screening experiments, medicinal chemistry studies, chemical biology research and drug discovery programs. In addition, the BioAssay database contains data from high-throughput RNA interference screening aimed at identifying critical genes responsible for a biological process or disease condition. The mission of PubChem is to serve the community by providing free and easy access to all deposited data. To this end, PubChem BioAssay is integrated into the National Center for Biotechnology Information retrieval system, making them searchable by Entrez queries and cross-linked to other biomedical information archived at National Center for Biotechnology Information. Moreover, PubChem BioAssay provides web-based and programmatic tools allowing users to search, access and analyze bioassay test results and metadata. In this work, we provide an update for the PubChem BioAssay resource, such as information content growth, new developments supporting data integration and search, and the recently deployed PubChem Upload to streamline chemical structure and bioassay submissions.

Computational design of protein-small molecule interfaces

The computational design of proteins that bind small molecule ligands is one of the unsolved challenges in protein engineering. It is complicated by the relatively small size of the ligand which limits the number of intermolecular interactions. Furthermore, near-perfect geometries between interacting partners are required to achieve high binding affinities. For apolar, rigid small molecules the interactions are dominated by short-range van der Waals forces. As the number of polar groups in the ligand increases, hydrogen bonds, salt bridges, cation-π, and π-π interactions gain importance. These partial covalent interactions are longer ranged, and additionally, their strength depends on the environment (e.g. solvent exposure). To assess the current state of protein-small molecule interface design, we benchmark the popular computer algorithm Rosetta on a diverse set of 43 protein-ligand complexes. On average, we achieve sequence recoveries in the binding site of 59% when the ligand is allowed limited reorientation, and 48% when the ligand is allowed full reorientation. When simulating the redesign of a protein binding site, sequence recovery among residues that contribute most to binding was 52% when slight ligand reorientation was allowed, and 27% when full ligand reorientation was allowed. As expected, sequence recovery correlates with ligand displacement.