Rules for Identifying Potentially Reactive or Promiscuous Compounds

Small molecule inhibitors targeting heat shock protein 90: An updated review

As a molecular chaperone, heat shock protein 90 (HSP90) plays important roles in the folding, stabilization, activation, and degradation of over 500 client proteins, and is extensively involved in cell signaling, proliferation, and survival. Thus, it has emerged as an important target in a variety of diseases, including cancer, neurodegenerative diseases, and viral infections. Therefore, targeted inhibition of HSP90 provides a valuable and promising therapeutic strategy for the treatment of HSP90-related diseases. This review aims to systematically summarize the progress of research on HSP90 inhibitors in the last five years, focusing on their structural features, design strategies, and biological activities. It will refer to the natural products and their derivatives (including novobiocin derivatives, deguelin derivatives, quinone derivatives, and terpenoid derivatives), and to synthetic small molecules (including resorcinol derivatives, pyrazoles derivatives, triazole derivatives, pyrimidine derivatives, benzamide derivatives, benzothiazole derivatives, and benzofuran derivatives). In addition, the major HSP90 small-molecule inhibitors that have moved into clinical trials to date are also presented here.

Small-Molecule Disruptors of the Interaction between Calcium- and Integrin-Binding Protein 1 and Integrin αIIbβ3 as Novel Antiplatelet Agents

Thrombosis, a key factor in most cardiovascular diseases, is a major contributor to human mortality. Existing antithrombotic agents carry a risk of bleeding. Consequently, there is a keen interest in discovering innovative antithrombotic agents that can prevent thrombosis without negatively impacting hemostasis. Platelets play crucial roles in both hemostasis and thrombosis. We have previously characterized calcium- and integrin-binding protein 1 (CIB1) as a key regulatory molecule that regulates platelet function. CIB1 interacts with several platelet proteins including integrin αIIbβ3, the major glycoprotein receptor for fibrinogen on platelets. Given that CIB1 regulates platelet function through its interaction with αIIbβ3, we developed a fluorescence polarization (FP) assay to screen for potential inhibitors. The assay was miniaturized to 1536-well and screened in quantitative high-throughput screening (qHTS) format against a diverse compound library of 14,782 compounds. After validation and selectivity testing using the FP assay, we identified 19 candidate inhibitors and validated them using an in-gel binding assay that monitors the interaction of CIB1 with αIIb cytoplasmic tail peptide, followed by testing of top hits by intrinsic tryptophan fluorescence (ITF) and microscale thermophoresis (MST) to ascertain their interaction with CIB1. Two of the validated hits shared similar chemical structures, suggesting a common mechanism of action. Docking studies further revealed promising interactions within the hydrophobic binding pocket of the target protein, particularly forming key hydrogen bonds with Ser180. The compounds exhibited a potent antiplatelet activity based on their inhibition of thrombin-induced human platelet aggregation, thus indicating that disruptors of the CIB1- αIIbβ3 interaction could carry a translational potential as antithrombotic agents.

ChemFH: an integrated tool for screening frequent false positives in chemical biology and drug discovery

High-throughput screening rapidly tests an extensive array of chemical compounds to identify hit compounds for specific biological targets in drug discovery. However, false-positive results disrupt hit compound screening, leading to wastage of time and resources. To address this, we propose ChemFH, an integrated online platform facilitating rapid virtual evaluation of potential false positives, including colloidal aggregators, spectroscopic interference compounds, firefly luciferase inhibitors, chemical reactive compounds, promiscuous compounds, and other assay interferences. By leveraging a dataset containing 823 391 compounds, we constructed high-quality prediction models using multi-task directed message-passing network (DMPNN) architectures combining uncertainty estimation, yielding an average AUC value of 0.91. Furthermore, ChemFH incorporated 1441 representative alert substructures derived from the collected data and ten commonly used frequent hitter screening rules. ChemFH was validated with an external set of 75 compounds. Subsequently, the virtual screening capability of ChemFH was successfully confirmed through its application to five virtual screening libraries. Furthermore, ChemFH underwent additional validation on two natural products and FDA-approved drugs, yielding reliable and accurate results. ChemFH is a comprehensive, reliable, and computationally efficient screening pipeline that facilitates the identification of true positive results in assays, contributing to enhanced efficiency and success rates in drug discovery. ChemFH is freely available via https://chemfh.scbdd.com/.

Tackling assay interference associated with small molecules

Biochemical and cell-based assays are essential to discovering and optimizing efficacious and safe drugs, agrochemicals and cosmetics. However, false assay readouts stemming from colloidal aggregation, chemical reactivity, chelation, light signal attenuation and emission, membrane disruption, and other interference mechanisms remain a considerable challenge in screening synthetic compounds and natural products. To address assay interference, a range of powerful experimental approaches are available and in silico methods are now gaining traction. This Review begins with an overview of the scope and limitations of experimental approaches for tackling assay interference. It then focuses on theoretical methods, discusses strategies for their integration with experimental approaches, and provides recommendations for best practices. The Review closes with a summary of the critical facts and an outlook on potential future developments.

The artificial neural network selects saccharides from natural sources a promise for potential FimH inhibitor to prevent UTI infections

The major challenge in the development of affordable medicines from natural sources is the unavailability of logical protocols to explain their mechanism of action in biological targets. FimH (Type 1 fimbrin with D-mannose specific adhesion property), a lectin on E. coli cell surface is a promising target to combat the urinary tract infection (UTI). The present study aimed at predicting the inhibitory capacity of saccharides on FimH. As mannosides are considered FimH inhibitors, the readily accessible saccharides from the PubChem collection were utilized. The artificial neural networks (ANN)-based machine learning algorithm Self-organizing map (SOM) has been successfully employed in predicting active molecules as they could discover relationships through self-organization for the ligand-based virtual screening. Docking was used for the structure-based virtual screening and molecular dynamic simulation for validation. The result revealed that the predicted molecules malonyl hexose and mannosyl glucosyl glycerate exhibit exactly similar binding interactions and better docking scores as that of the reference bioassay active, heptyl mannose. The pharmacokinetic profile matches that of the selected bioflavonoids (quercetin malonyl hexose, kaempferol malonyl hexose) and has better values than the control drug bioflavonoid, monoxerutin. Thus, these two molecules can effectively inhibit type 1 fimbrial adhesin, as antibiotics against E. coli and can be explored as a prophylactic against UTIs. Moreover, this investigation can pave the way to the exploration of the potential benefits of plant-based treatments.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s40203-024-00212-5.

Abiotic Small Molecule Inhibitors and Activators of the LasR Quorum Sensing Receptor in Pseudomonas aeruginosa with Potencies Comparable or Surpassing N-Acyl Homoserine Lactones

The opportunistic pathogen Pseudomonas aeruginosa controls almost 10% of its genome, including myriad virulence genes, via a cell-to-cell chemical communication system called quorum sensing (QS). Small molecules that either inhibit or activate QS in P. aeruginosa represent useful research tools to study the role of this signaling pathway in infection and interrogate its viability as an antivirulence target. However, despite active research in this area over the past 20+ years, there are relatively few synthetic compounds known to strongly inhibit or activate QS in P. aeruginosa. Most reported QS modulators in this pathogen are of low potency or have structural liabilities that limit their application in biologically relevant environments such as mimics of the native N-acyl l-homoserine lactone (AHL) signals. Here, we report the results of a high-throughput screen for abiotic small molecules that target LasR, a key QS regulator in P. aeruginosa. We screened a 25,000-compound library and discovered four new structural classes of abiotic LasR modulators. These compounds include antagonists that surpass the potency of all known AHL-type compounds and mimetics thereof, along with an agonist with potency approaching that of LasR's native ligand. The novel structures of this compound set, along with their anticipated robust physicochemical profiles, underscore their potential value as probe molecules to interrogate the roles of QS in this formidable pathogen.

AI is a viable alternative to high throughput screening: a 318-target study

High throughput screening (HTS) is routinely used to identify bioactive small molecules. This requires physical compounds, which limits coverage of accessible chemical space. Computational approaches combined with vast on-demand chemical libraries can access far greater chemical space, provided that the predictive accuracy is sufficient to identify useful molecules. Through the largest and most diverse virtual HTS campaign reported to date, comprising 318 individual projects, we demonstrate that our AtomNet® convolutional neural network successfully finds novel hits across every major therapeutic area and protein class. We address historical limitations of computational screening by demonstrating success for target proteins without known binders, high-quality X-ray crystal structures, or manual cherry-picking of compounds. We show that the molecules selected by the AtomNet® model are novel drug-like scaffolds rather than minor modifications to known bioactive compounds. Our empirical results suggest that computational methods can substantially replace HTS as the first step of small-molecule drug discovery.

Relationship between prediction accuracy and uncertainty in compound potency prediction using deep neural networks and control models

The assessment of prediction variance or uncertainty contributes to the evaluation of machine learning models. In molecular machine learning, uncertainty quantification is an evolving area of research where currently no standard approaches or general guidelines are available. We have carried out a detailed analysis of deep neural network variants and simple control models for compound potency prediction to study relationships between prediction accuracy and uncertainty. For comparably accurate predictions obtained with models of different complexity, highly variable prediction uncertainties were detected using different metrics. Furthermore, a strong dependence of prediction characteristics and uncertainties on potency levels of test compounds was observed, often leading to over- or under-confident model decisions with respect to the expected variance of predictions. Moreover, neural network models responded very differently to training set modifications. Taken together, our findings indicate that there is only little, if any correlation between compound potency prediction accuracy and uncertainty, especially for deep neural network models, when predictions are assessed on the basis of currently used metrics for uncertainty quantification.

Comprehensive Data-Driven Assessment of Non-Kinase Targets of Inhibitors of the Human Kinome

Protein kinases (PKs) are involved in many intracellular signal transduction pathways through phosphorylation cascades and have become intensely investigated pharmaceutical targets over the past two decades. Inhibition of PKs using small-molecular inhibitors is a premier strategy for the treatment of diseases in different therapeutic areas that are caused by uncontrolled PK-mediated phosphorylation and aberrant signaling. Most PK inhibitors (PKIs) are directed against the ATP cofactor binding site that is largely conserved across the human kinome comprising 518 wild-type PKs (and many mutant forms). Hence, these PKIs often have varying degrees of multi-PK activity (promiscuity) that is also influenced by factors such as single-site mutations in the cofactor binding region, compound binding kinetics, and residence times. The promiscuity of PKIs is often-but not always-critically important for therapeutic efficacy through polypharmacology. Various in vitro and in vivo studies have also indicated that PKIs have the potential of interacting with additional targets other than PKs, and different secondary cellular targets of individual PKIs have been identified on a case-by-case basis. Given the strong interest in PKs as drug targets, a wealth of PKIs from medicinal chemistry and their activity data from many assays and biological screens have become publicly available over the years. On the basis of these data, for the first time, we conducted a systematic search for non-PK targets of PKIs across the human kinome. Starting from a pool of more than 155,000 curated human PKIs, our large-scale analysis confirmed secondary targets from diverse protein classes for 447 PKIs on the basis of high-confidence activity data. These PKIs were active against 390 human PKs, covering all kinase groups of the kinome and 210 non-PK targets, which included other popular pharmaceutical targets as well as currently unclassified proteins. The target distribution and promiscuity of the 447 PKIs were determined, and different interaction profiles with PK and non-PK targets were identified. As a part of our study, the collection of PKIs with activity against non-PK targets and the associated information are made freely available.

Curation and cheminformatics analysis of a Ugi-reaction derived library (URDL) of synthetically tractable small molecules for virtual screening application

Virtual screening (VS) is an important approach in drug discovery and relies on the availability of a virtual library of synthetically tractable molecules. Ugi reaction (UR) represents an important multi-component reaction (MCR) that reliably produces a peptidomimetic scaffold. Recent literature shows that a tactically assembled Ugi adduct can be subjected to further chemical modifications to yield a variety of rings and scaffolds, thus, renewing the interest in this old reaction. Given the reliability and efficiency of UR, we collated an UR derived library (URDL) of small molecules (total = 5773) for VS. The synthesis of the majority of URDL molecules may be carried out in 1-2 pots in a time and cost-effective manner. The detailed analysis of the average property and chemical space of URDL was also carried out using the open-source Datawarrior program. The comparison with FDA-approved oral drugs and inhibitors of protein-protein interactions (iPPIs) suggests URDL molecules are 'clean', drug-like, and conform to a structurally distinct space from the other two categories. The average physicochemical properties of compounds in the URDL library lie closer to iPPI molecules than oral drugs thus suggesting that the URDL resource can be applied to discover novel iPPI molecules. The URDL molecules consist of diverse ring systems, many of which have not been exploited yet for drug design. Thus, URDL represents a small virtual library of drug-like molecules with unexplored chemical space designed for VS. The structures of all molecules of URDL, oral drugs, and iPPI compounds are being made freely accessible as supplementary information for broader application.

Atypical N-Alkyl to N-Noralkoxy Switch in a Dual cSRC/BCR-ABL1 Kinase Inhibitor Improves Drug Efflux and hERG Affinity

We describe an atypical amine bioisostere, the trisubstituted hydroxylamine, that upon incorporation into an approved dual cSRC/BCR-ABL1 kinase inhibitor yields 9, a compound that retains potent biological activity and couples it with improved drug efflux and hERG affinity at the expense of only a 2 atomic mass unit increase in molecular weight. Contrary to the common expectation for hydroxylamines in medicinal chemistry, 9 is well tolerated in vivo and lacks the mutagenicity and genotoxicity so often ascribed to lesser substituted hydroxylamines. A matched molecular pair (MMP) analysis suggests that the beneficial properties conferred by the N-alkyl to N-noralkoxy switch arises from a reduction in basicity of the piperazine unit. Overall, these results lend additional support to the use of trisubstituted hydroxylamines as bioisosteres of N-alkyl groups that are not involved in key polar interactions.

Artificial intelligence: Machine learning approach for screening large database and drug discovery

Recent research in drug discovery dealing with many faces difficulties, including development of new drugs during disease outbreak and drug resistance due to rapidly accumulating mutations. Virtual screening is the most widely used method in computer aided drug discovery. It has a prominent ability in screening drug targets from large molecular databases. Recently, a number of web servers have developed for quickly screening publicly accessible chemical databases. In a nutshell, deep learning algorithms and artificial neural networks have modernised the field. Several drug discovery processes have used machine learning and deep learning algorithms, including peptide synthesis, structure-based virtual screening, ligand-based virtual screening, toxicity prediction, drug monitoring and release, pharmacophore modelling, quantitative structure-activity relationship, drug repositioning, polypharmacology, and physiochemical activity. Although there are presently a wide variety of data-driven AI/ML tools available, the majority of these tools have, up to this point, been developed in the context of non-communicable diseases like cancer, and a number of obstacles have prevented the translation of these tools to the discovery of treatments against infectious diseases. In this review various aspects of AI and ML in virtual screening of large databases were discussed. Here, with an emphasis on antivirals as well as other disease, offers a perspective on the advantages, drawbacks, and hazards of AI/ML techniques in the search for innovative treatments.

Anatomy of Potency Predictions Focusing on Structural Analogues with Increasing Potency Differences Including Activity Cliffs

Potency predictions are popular in compound design and optimization but are complicated by intrinsic limitations. Moreover, even for nonlinear methods, activity cliffs (ACs, formed by structural analogues with large potency differences) represent challenging test cases for compound potency predictions. We have devised a new test system for potency predictions, including AC compounds, that is based on partitioned matched molecular pairs (MMP) and makes it possible to monitor prediction accuracy at the level of analogue pairs with increasing potency differences. The results of systematic predictions using different machine learning and control methods on MMP-based data sets revealed increasing prediction errors when potency differences between corresponding training and test compounds increased, including large prediction errors for AC compounds. At the global level, these prediction errors were not apparent due to the statistical dominance of analogue pairs with small potency differences. Test compounds from such pairs were accurately predicted and determined the observed global prediction accuracy. Shapley value analysis, an explainable artificial intelligence approach, was applied to identify structural features determining potency predictions using different methods. The analysis revealed that numerical predictions of different regression models were determined by features that were shared by MMP partner compounds or absent in these compounds, with opposing effects. These findings provided another rationale for accurate predictions of similar potency values for structural analogues and failures in predicting the potency of AC compounds.

Discovery of a Hydroxylamine-Based Brain-Penetrant EGFR Inhibitor for Metastatic Non-Small-Cell Lung Cancer

Metastases to the brain remain a significant problem in lung cancer, as treatment by most small-molecule targeted therapies is severely limited by efflux transporters at the blood-brain barrier (BBB). Here, we report the discovery of a selective, orally bioavailable, epidermal growth factor receptor (EGFR) inhibitor, 9, that exhibits high brain penetration and potent activity in osimertinib-resistant cell lines bearing L858R/C797S and exon19del/C797S EGFR resistance mutations. In vivo, 9 induced tumor regression in an intracranial patient-derived xenograft (PDX) murine model suggesting it as a potential lead for the treatment of localized and metastatic non-small-cell lung cancer (NSCLC) driven by activating mutant bearing EGFR. Overall, we demonstrate that an underrepresented functional group in medicinal chemistry, the trisubstituted hydroxylamine moiety, can be incorporated into a drug scaffold without the toxicity commonly surmised to accompany these units, all while maintaining potent biological activity and without the molecular weight creep common to drug optimization campaigns.

Identification of inhibitors targeting the energy-coupling factor (ECF) transporters

The energy-coupling factor (ECF) transporters are a family of transmembrane proteins involved in the uptake of vitamins in a wide range of bacteria. Inhibition of the activity of these proteins could reduce the viability of pathogens that depend on vitamin uptake. The central role of vitamin transport in the metabolism of bacteria and absence from humans make the ECF transporters an attractive target for inhibition with selective chemical probes. Here, we report on the identification of a promising class of inhibitors of the ECF transporters. We used coarse-grained molecular dynamics simulations on Lactobacillus delbrueckii ECF-FolT2 and ECF-PanT to profile the binding mode and mechanism of inhibition of this novel chemotype. The results corroborate the postulated mechanism of transport and pave the way for further drug-discovery efforts.

Rationalizing general limitations in assessing and comparing methods for compound potency prediction

Compound potency predictions play a major role in computational drug discovery. Predictive methods are typically evaluated and compared in benchmark calculations that are widely applied. Previous studies have revealed intrinsic limitations of potency prediction benchmarks including very similar performance of increasingly complex machine learning methods and simple controls and narrow error margins separating machine learning from randomized predictions. However, origins of these limitations are currently unknown. We have carried out an in-depth analysis of potential reasons leading to artificial outcomes of potency predictions using different methods. Potency predictions on activity classes typically used in benchmark settings were found to be determined by compounds with intermediate potency close to median values of the compound data sets. The potency of these compounds was consistently predicted with high accuracy, without the need for learning, which dominated the results of benchmark calculations, regardless of the activity classes used. Taken together, our findings provide a clear rationale for general limitations of compound potency benchmark predictions and a basis for the design of alternative test systems for methodological comparisons.

Exploring the Translational Gap of a Novel Class of Escherichia coli IspE Inhibitors

Discovery of novel antibiotics needs multidisciplinary approaches to gain target enzyme and bacterial activities while aiming for selectivity over mammalian cells. Here, we report a multiparameter optimisation of a fragment-like hit that was identified through a structure-based virtual-screening campaign on Escherichia coli IspE crystal structure. Subsequent medicinal-chemistry design resulted in a novel class of E. coli IspE inhibitors, exhibiting activity also against the more pathogenic bacteria Pseudomonas aeruginosa and Acinetobacter baumannii. While cytotoxicity remains a challenge for the series, it provides new insights on the molecular properties for balancing enzymatic target and bacterial activities simultaneously as well as new starting points for the development of IspE inhibitors with a predicted new mode of action.

Pharmacophore modeling, molecular docking, and molecular dynamics studies to identify new 5-HT2AR antagonists with the potential for design of new atypical antipsychotics

Some important atypical antipsychotic drugs target the serotonergic receptor 2A (5-HT2AR). Currently, new therapeutic strategies are needed to offer faster onset of action with fewer side effects and, therefore, greater efficacy in a substantial proportion of patients with neuropsychological disorders such as Autism and Parkinson. The main objective of this work was to use SBDD methods to identify new hit compounds potentially useful as precursors of novel and selective 5-HT2AR antagonists. A structure-based pharmacophore screening study based on a selective antagonist was carried out in ten databases. The set obtained was refined using molecular docking, and the five most promising compounds were subjected to molecular dynamics simulations. The most stable and promising hit occupied a side pocket present in the 5-HT2AR, a site that can be explored to obtain selective ligands. Simulations against 5-HT2CR and D2R showed that the best hit could not form stable complexes with these targets, strengthening the hypothesis that the hit presents selective binding by the receptor of interest. The selected hits showed some predicted toxicity risk or violated some drug-likeness property. However, it can be concluded that the identified hits are the most promising for performing in vitro assays. Once the presence of activity is confirmed, they could become precursors of optimized and selective antagonists of 5-HT2AR. An SBDD study was carried out to identify new selective 5-HT2AR ligands potentially useful for designing selective atypical antipsychotics.

Silver Nanoparticles from Saudi and Syrian Black Cumin Seed Extracts: Green Synthesis, ADME, Toxicity, Comparative Research, and Biological Appraisal

Objective

The current study's objective is to highlight the value of using plant resources to identify key bioactive molecules and implement green chemistry in research and development to meet market demand.

Materials and Methods

The black cumin seeds (Saudi and Syria originated) were utilized to make silver nanoparticles (Ag-NPs), which were subsequently confirmed using a UV spectrophotometer and color analysis of reaction mixtures. The antibacterial activity of Ag-NPs was tested against E. coli, K. pneumoniae, and S. aureus, and antioxidant activity was measured using the DPPH assay. Swiss-ADME, pkCSM, and ProTox-II were also used to assess the pharmacokinetics, oral bioavailability, toxicity, and safety endpoints of molecules.

Result

The antibacterial effect of Ag-NPs from Saudi-origin black cumin seeds was observed higher. In comparison to the standard, the Saudi and Syrian Ag-NPs combined displayed synergistic antibacterial effects and were found to be more susceptible to S. aureus. In comparison to the reference, the antioxidant activity of Ag-NPs indicated 60-85% radical scavenging. All molecules passed the Lipinski rule, the filter (Veber, Egan, and Muegge), PAINS, and the Brenk structural alert (zero violations), and the synthetic score was also found to be in the easy limit (1 to 2). The compounds were found to be non-substrate for p-glycoprotein, high GIA% (>90%), non-inhibitor for CYP3A4, CYP2C19, CYP2C9, CYP2D6 (except 5 and 10), Log Po/w (1.71 to 3.26), TPSA 150 2 and MR 155. The compounds likewise had high Caco2 values (log Papp >0.9) with the exception of 4 and 9 (log Papp 0.9), were non-inhibitors of P-gp-I and II and hERG I and II, and showed no AMES toxicity. Except for molecule 11, no organ damage (hepatotoxicity) or endpoint toxicity (mutagenicity, immunotoxicity, carcinogenicity, and cytotoxicity) was identified in ProTox-II.

Conclusion

The current study sheds new light on the significance of bioactive molecules found in black cumin seeds, with molecules 3 and 6 identified as potential leads (highest GIA%, no AMES toxicity, oral rat acute and chronic toxicity, lack of renal OCT2 substrate, high total clearance, and lack of organ toxicity) for further research for a variety of medical applications.

DIY Virtual Chemical Libraries - Novel Starting Points for Drug Discovery

The advancement of in silico technologies such as library enumeration and synthetic feasibility prediction has made drug discovery pipelines rely more and more on virtual libraries, which provide a significantly larger pool of compounds than in-stock supplier catalogs. Virtual libraries from external sources, however, may be associated with long delivery time and high cost. In this study, we present a Do-It-Yourself (DIY) combinatorial chemistry library containing over 14 million almost completely novel products built from 1000 low-cost building blocks based on robust reactions frequently applied at medicinal chemistry laboratories. The applicability of the DIY library for various drug discovery approaches is demonstrated by extensive physicochemical property, structural diversity profiling, and the generation of focused libraries. We found that internally built DIY chemical libraries present a viable alternative of external virtual catalogs by providing access to a large number of low-cost and quickly accessible potential chemical starting points for drug discovery.

Explaining Multiclass Compound Activity Predictions Using Counterfactuals and Shapley Values

Most machine learning (ML) models produce black box predictions that are difficult, if not impossible, to understand. In pharmaceutical research, black box predictions work against the acceptance of ML models for guiding experimental work. Hence, there is increasing interest in approaches for explainable ML, which is a part of explainable artificial intelligence (XAI), to better understand prediction outcomes. Herein, we have devised a test system for the rationalization of multiclass compound activity prediction models that combines two approaches from XAI for feature relevance or importance analysis, including counterfactuals (CFs) and Shapley additive explanations (SHAP). For compounds with different single- and dual-target activities, we identified small compound modifications that induce feature changes inverting class label predictions. In combination with feature mapping, CFs and SHAP value calculations provide chemically intuitive explanations for model decisions.

Discovery and development of botanical natural products and their analogues as therapeutics for ovarian cancer

Covering: 2015 through the end of July 2022Ovarian cancer is one of the most common cancers affecting the female reproductive organs and has the highest mortality rate among gynecological cancers. Although botanical drugs and their derivatives, namely members of the taxane and camptothecin families, represent significant therapeutics currently available for the treatment of ovarian cancer, new drugs that have alternative mechanisms of action are still needed to combat the disease. For this reason, many efforts to identify additional novel compounds from botanical sources, along with the further development of existing therapeutics, have continued to appear in the literature. This review is designed to serve as a comprehensive look at both the currently available small-molecule therapeutic options and the recently reported botanically-derived natural products currently being studied and developed as potential future therapeutics that could one day be used against ovarian cancer. Specifically, key properties, structural features, and biological data are highlighted that are important for the successful development of potential agents. Recently reported examples are specifically discussed in the context of "drug discovery attributes," including the presence of structure-activity relationship, mechanism of action, toxicity, and pharmacokinetic studies, to help indicate the potential for future development and to highlight where these compounds currently exist in the development process. The lessons learned from both the successful development of the taxanes and camptothecins, as well as the strategies currently being employed for new drug development, are expected to ultimately help guide the future development of botanical natural products for ovarian cancer.

Integrated data-driven and experimental approaches to accelerate lead optimization targeting SARS-CoV-2 main protease

Identification of potential therapeutic candidates can be expedited by integrating computational modeling with domain aware machine learning (ML) models followed by experimental validation in an iterative manner. Generative deep learning models can generate thousands of new candidates, however, their physiochemical and biochemical properties are typically not fully optimized. Using our recently developed deep learning models and a scaffold as a starting point, we generated tens of thousands of compounds for SARS-CoV-2 M^pro that preserve the core scaffold. We utilized and implemented several computational tools such as structural alert and toxicity analysis, high throughput virtual screening, ML-based 3D quantitative structure-activity relationships, multi-parameter optimization, and graph neural networks on generated candidates to predict biological activity and binding affinity in advance. As a result of these combined computational endeavors, eight promising candidates were singled out and put through experimental testing using Native Mass Spectrometry and FRET-based functional assays. Two of the tested compounds with quinazoline-2-thiol and acetylpiperidine core moieties showed IC[Formula: see text] values in the low micromolar range: [Formula: see text] [Formula: see text]M and 3.41±0.0015 [Formula: see text]M, respectively. Molecular dynamics simulations further highlight that binding of these compounds results in allosteric modulations within the chain B and the interface domains of the M^pro. Our integrated approach provides a platform for data driven lead optimization with rapid characterization and experimental validation in a closed loop that could be applied to other potential protein targets.

Thiocarbonyl Derivatives of Natural Chlorins: Synthesis Using Lawesson's Reagent and a Study of Their Properties

The development of sulfur-containing pharmaceutical compounds is important in the advancement of medicinal chemistry. Photosensitizers (PS) that acquire new properties upon incorporation of sulfur-containing groups or individual sulfur atoms into their structure are not neglected, either. In this work, a synthesis of sulfur-containing derivatives of natural chlorophyll a using Lawesson's reagent was optimized. Thiocarbonyl chlorins were shown to have a significant bathochromic shift in the absorption and fluorescence bands. The feasibility of functionalizing the thiocarbonyl group at the macrocycle periphery by formation of a Pt(II) metal complex in the chemotherapeutic agent cisplatin was shown. The chemical stability of the resulting conjugate in aqueous solution was studied, and it was found to possess a high cytotoxic activity against sarcoma S37 tumor cells that results from the combined photodynamic and chemotherapeutic effect on these cells.

Fragtory: Pharmacophore-Focused Design, Synthesis, and Evaluation of an sp3-Enriched Fragment Library

Fragment-based drug discovery has played an important role in medicinal chemistry and pharmaceutical research. Despite numerous demonstrated successes, the limited diversity and overrepresentation of planar, sp²-rich structures in commercial libraries often hamper the full potential of this approach. Hence, the thorough design of screening libraries inevitably determines the probability for meaningful hits and subsequent structural elaboration. Against this background, we present the generation of an exclusive fragment library based on iterative entry nomination by a specifically designed computational workflow: "Fragtory". Following a pharmacophore diversity-driven approach, we used Fragtory in an interdisciplinary academic setting to guide both tailored synthesis efforts and the implementation of in-house compounds to build a curated 288-member library of sp³-enriched fragments. Subsequent NMR screens against a model protein and hit validation by protein crystallography led to the identification of structurally novel ligands that were further characterized by isothermal titration calorimetry, demonstrating the applicability of our experimental approach.

Differences in learning characteristics between support vector machine and random forest models for compound classification revealed by Shapley value analysis

The random forest (RF) and support vector machine (SVM) methods are mainstays in molecular machine learning (ML) and compound property prediction. We have explored in detail how binary classification models derived using these algorithms arrive at their predictions. To these ends, approaches from explainable artificial intelligence (XAI) are applicable such as the Shapley value concept originating from game theory that we adapted and further extended for our analysis. In large-scale activity-based compound classification using models derived from training sets of increasing size, RF and SVM with the Tanimoto kernel produced very similar predictions that could hardly be distinguished. However, Shapley value analysis revealed that their learning characteristics systematically differed and that chemically intuitive explanations of accurate RF and SVM predictions had different origins.

Large-Scale Predictions of Compound Potency with Original and Modified Activity Classes Reveal General Prediction Characteristics and Intrinsic Limitations of Conventional Benchmarking Calculations

Predicting compound potency is a major task in computational medicinal chemistry, for which machine learning is often applied. This study systematically predicted compound potency values for 367 target-based compound activity classes from medicinal chemistry using a preferred machine learning approach and simple control methods. The predictions produced unexpectedly similar results for different classes and comparably high accuracy for machine learning and simple control models. Based on these findings, the influence of different data set modifications on relative prediction accuracies was explored, including potency range balancing, removal of nearest neighbors, and analog series-based compound partitioning. The predictions were surprisingly resistant to these modifications, leading to only small error margin increases. These findings also show that conventional benchmark settings are unsuitable for directly comparing potency prediction methods.

Recent Advances in the Synthesis of Di- and Trisubstituted Hydroxylamines

As an underrepresented functional group in bioorganic and medicinal chemistry, the hydroxylamine unit has historically received little attention from the synthetic community. Recent developments, however, suggest that hydroxylamines may have broader applications such that a review covering recent developments in the synthesis of this functional group is timely. With this in mind, this review primarily covers developments in the past 15 years in the preparation of di- and trisubstituted hydroxylamines. The mechanism of the reactions and key features and shortcomings are discussed throughout the review.

Exhaustive in silico design and screening of novel antipsychotic compounds with improved pharmacodynamics and blood-brain barrier permeation properties

Antipsychotic drugs or neuroleptics are widely used in the treatment of psychosis as a manifestation of schizophrenia and bipolar disorder. However, their effectiveness largely depends on the blood-brain barrier (BBB) permeation (pharmacokinetics) and drug-receptor pharmacodynamics. Therefore, in this study, we developed and implemented the in silico pipeline to design novel compounds (n = 260) as leads using the standard drug scaffolds with improved PK/PD properties from the standard scaffolds. As a result, the best candidates (n = 3) were evaluated in molecular docking to interact with serotonin and dopamine receptors. Finally, haloperidol (HAL) derivative (1-(4-fluorophenyl)-4-(4-hydroxy-4-{4-[(2-phenyl-1,3-thiazol-4-yl)methyl]phenyl}piperidin-1-yl)butan-1-one) was identified as a "magic shotgun" lead compound with better affinity to the 5-HT2A, 5-HT1D, D2, D3, and 5-HT1B receptors than the control molecule. Additionally, this hit substance was predicted to possess similar BBB permeation properties and much lower toxicological profiles in comparison to HAL. Overall, the proposed rational drug design platform for novel antipsychotic drugs based on the BBB permeation and receptor binding might be an invaluable asset for a medicinal chemist or translational pharmacologist.Communicated by Ramaswamy H. Sarma.

The European Federation for Medicinal Chemistry and Chemical Biology (EFMC) Best Practice Initiative: Hit Generation

Hit generation is a crucial step in drug discovery that will determine the speed and chance of success of identifying drug candidates. Many strategies are now available to identify chemical starting points, or hits, and each biological target warrants a tailored approach. In this set of best practices, we detail the essential approaches for target centric hit generation and the opportunities and challenges they come with. We then provide guidance on how to validate hits to ensure medicinal chemistry is only performed on compounds and scaffolds that engage the target of interest and have the desired mode of action. Finally, we discuss the design of integrated hit generation strategies that combine several approaches to maximize the chance of identifying high quality starting points to ensure a successful drug discovery campaign.

Predicting Potent Compounds Using a Conditional Variational Autoencoder Based upon a New Structure-Potency Fingerprint

Prediction of the potency of bioactive compounds generally relies on linear or nonlinear quantitative structure-activity relationship (QSAR) models. Nonlinear models are generated using machine learning methods. We introduce a novel approach for potency prediction that depends on a newly designed molecular fingerprint (FP) representation. This structure-potency fingerprint (SPFP) combines different modules accounting for the structural features of active compounds and their potency values in a single bit string, hence unifying structure and potency representation. This encoding enables the derivation of a conditional variational autoencoder (CVAE) using SPFPs of training compounds and apply the model to predict the SPFP potency module of test compounds using only their structure module as input. The SPFP-CVAE approach correctly predicts the potency values of compounds belonging to different activity classes with an accuracy comparable to support vector regression (SVR), representing the state-of-the-art in the field. In addition, highly potent compounds are predicted with very similar accuracy as SVR and deep neural networks.

Synthesis, Characterization, and Pharmacokinetic Studies of Thiazolidine-2,4-Dione Derivatives

Various derivatives of thiazolidine-2,4-dione (C1–C5) were designed and synthesized by chemical reaction with 4-nitrobenzaldehyde using Knoevenagel reaction conditions which results in the reduction of nitro group to amine and further modification results in target compounds. The chemical structures of all the 2,4-thiazolidinedione derivatives have been elucidated by 1H and 13C NMR spectroscopy. These compounds were further characterized by in silico ADME (absorption, distribution, metabolism, and excretion) studies. The pharmacokinetic properties were assessed by SwissADME software. The in silico ADME (absorption, distribution, metabolism, and excretion) assessment reveals that all derivatives (C1 to C5) have 5 to 7 rotatable bonds. Lipophilicity and water solubility showed that C1, C2, and C4 are water soluble except for C3 and C5 which are moderately soluble. All the compounds have high GI absorption except C3. None of the derivatives are blood-brain barrier permeant. Drug metabolism of TZDs derivatives showed that C3 was identified as an inhibitor of CYP2C9 and C5 as an inhibitor of CYP1A2 and CYP2C19. Drug likeness properties indicate that C1 has only one violation of the Ghose rule while C3 has violations in the Ghose and Egan rules. The in silico pharmacokinetic studies revealed high GI absorption and the inability to pass blood-brain barrier which can be further assessed by in vitro and in vivo antihyperglycemic activity. This study will contribute to providing TZDs derivatives with an improved pharmacokinetic profile and decreased toxicity.

Explaining Accurate Predictions of Multitarget Compounds with Machine Learning Models Derived for Individual Targets

In drug discovery, compounds with well-defined activity against multiple targets (multitarget compounds, MT-CPDs) provide the basis for polypharmacology and are thus of high interest. Typically, MT-CPDs for polypharmacology have been discovered serendipitously. Therefore, over the past decade, computational approaches have also been adapted for the design of MT-CPDs or their identification via computational screening. Such approaches continue to be under development and are far from being routine. Recently, different machine learning (ML) models have been derived to distinguish between MT-CPDs and corresponding compounds with activity against the individual targets (single-target compounds, ST-CPDs). When evaluating alternative models for predicting MT-CPDs, we discovered that MT-CPDs could also be accurately predicted with models derived for corresponding ST-CPDs; this was an unexpected finding that we further investigated using explainable ML. The analysis revealed that accurate predictions of ST-CPDs were determined by subsets of structural features of MT-CPDs required for their prediction. These findings provided a chemically intuitive rationale for the successful prediction of MT-CPDs using different ML models and uncovered general-feature subset relationships between MT- and ST-CPDs with activities against different targets.

In silico ADMET analysis of the A-, B- and D-modified androstane derivatives with potential anticancer effects

The major challenge in the fight against cancer is to design new drugs that will be more selective for cancer cells, with fewer side effects. Synthetic steroids such as cyproterone, fulvestrant, exemestane and abiraterone are approved powerful drugs for the treatment of hormone-dependent diseases such as breast and prostate cancers. Therefore, androstane derivatives in 17-substituted, 17a-homo lactone and 16,17-seco series, with potent anticancer activity, were selected for pharmacokinetic and druglike predictions from the absorption, distribution, metabolism and excretion (ADME) models. In silico determination of physico-chemical and ADMET properties was performed using SwissADME and ProTox-II web tools. The possibility of gastrointestinal absorption and brain penetration was analyzed using the BOILED-Egg model, while the in silico evaluation of the similarities between selected steroid derivatives and FDA-approved drugs was carried out using the SwissSimilarity tool. Of all tested, two compounds that showed good in silico ADMET results, in addition to promising cytotoxicity and molecular docking results, could potentially be evaluated in in vivo tests.

Introducing ligand GA, a genetic algorithm molecular tool for automated protein inhibitor design

Ligand GA is introduced in this work and approaches the problem of finding small molecules inhibiting protein functions by using the protein site to find close to optimal or optimal small molecule binders. Genetic algorithms (GA) are an effective means for approximating or solving computationally hard mathematics problems with large search spaces such as this one. The algorithm is designed to include constraints on the generated molecules from ADME restriction, localization in a binding site, specified hydrogen bond requirements, toxicity prevention from multiple proteins, sub-structure restrictions, and database inclusion. This algorithm and work is in the context of computational modeling, ligand design and docking to protein sites.

In Silico Identification of Multi-Target Ligands as Promising Hit Compounds for Neurodegenerative Diseases Drug Development

The conventional treatment of neurodegenerative diseases (NDDs) is based on the "one molecule-one target" paradigm. To combat the multifactorial nature of NDDs, the focus is now shifted toward the development of small-molecule-based compounds that can modulate more than one protein target, known as "multi-target-directed ligands" (MTDLs), while having low affinity for proteins that are irrelevant for the therapy. The in silico approaches have demonstrated a potential to be a suitable tool for the identification of MTDLs as promising drug candidates with reduction in cost and time for research and development. In this study more than 650,000 compounds were screened by a series of in silico approaches to identify drug-like compounds with predicted activity simultaneously towards three important proteins in the NDDs symptomatic treatment: acetylcholinesterase (AChE), histone deacetylase 2 (HDAC2), and monoamine oxidase B (MAO-B). The compounds with affinities below 5.0 µM for all studied targets were additionally filtered to remove known non-specifically binding or unstable compounds. The selected four hits underwent subsequent refinement through in silico blood-brain barrier penetration estimation, safety evaluation, and molecular dynamics simulations resulting in two hit compounds that constitute a rational basis for further development of multi-target active compounds against NDDs.

Fragment screening at AstraZeneca: developing the next generation biophysics fragment set

Fragment based drug discovery is a critical part of the lead generation toolbox and relies heavily on a readily available, high quality fragment library. Over years of use, the AstraZeneca fragment set had become partially depleted and instances of compound deterioration had been found. It was recognised that a redevelopment was required. This provided an opportunity to evolve our screening sets strategy, whilst ensuring that the quality of the fragment set met the robust requirements of fragment screening campaigns. In this communication we share the strategy employed, in particular highlighting two aspects of our approach that we believe others in the community would benefit from, namely that; (i) fragments were selected with input from Medicinal Chemists at an early stage, and (ii) the library was arranged in a layered format to ensure maximum flexibility on a per target basis.

An Insight into Symmetrical Cyanine Dyes as Promising Selective Antiproliferative Agents in Caco-2 Colorectal Cancer Cells

Cancer remains one of the diseases with the highest worldwide incidence. Several cytotoxic approaches have been used over the years to overcome this public health threat, such as chemotherapy, radiotherapy, and photodynamic therapy (PDT). Cyanine dyes are a class of compounds that have been extensively studied as PDT sensitisers; nevertheless, their antiproliferative potential in the absence of a light source has been scarcely explored. Herein, the synthesis of eighteen symmetric mono-, tri-, and heptamethine cyanine dyes and their evaluation as potential anticancer agents is described. The influences of the heterocyclic nature, counterion, and methine chain length on the antiproliferative effects and selectivities were analysed, and relevant structure-activity relationship data were gathered. The impact of light on the cytotoxic activity of the most promising dye was also assessed and discussed. Most of the monomethine and trimethine cyanine dyes under study demonstrated a high antiproliferative effect on human tumour cell lines of colorectal (Caco-2), breast (MCF-7), and prostate (PC-3) cancer at the initial screening (10 µM). However, concentration-viability curves showed higher potency and selectivity for the Caco-2 cell line. A monomethine cyanine dye derived from benzoxazole was the most promising compound (IC₅₀ for Caco-2 = 0.67 µM and a selectivity index of 20.9 for Caco-2 versus normal human dermal fibroblasts (NHDF)) and led to Caco-2 cell cycle arrest at the G0/G1 phase. Complementary in silico studies predicted good intestinal absorption and oral bioavailability for this cyanine dye.

EdgeSHAPer: Bond-Centric Shapley Value-Based Explanation Method for Graph Neural Networks

Graph neural networks (GNNs) recursively propagate signals along the edges of an input graph, integrate node feature information with graph structure, and learn object representations. Like other deep neural network models, GNNs have notorious black box character. For GNNs, only few approaches are available to rationalize model decisions. We introduce EdgeSHAPer, a generally applicable method for explaining GNN-based models. The approach is devised to assess edge importance for predictions. Therefore, EdgeSHAPer makes use of the Shapley value concept from game theory. For proof-of-concept, EdgeSHAPer is applied to compound activity prediction, a central task in drug discovery. EdgeSHAPer’s edge centricity is relevant for molecular graphs where edges represent chemical bonds. Combined with feature mapping, EdgeSHAPer produces intuitive explanations for compound activity predictions. Compared to a popular node-centric and another edge-centric GNN explanation method, EdgeSHAPer reveals higher resolution in differentiating features determining predictions and identifies minimal pertinent positive feature sets.

•

EdgeSHAPer is new methodology for explaining graph neural network models

•

Edge centricity represents a characteristic feature of the approach

•

EdgeSHAPer is generally applicable including molecular predictions

•

EdgeSHAPer produces explanations of compound predictions at a high resolution

Calculation of Exact Shapley Values for Support Vector Machines with Tanimoto Kernel Enables Model Interpretation

The support vector machine (SVM) algorithm is popular in chemistry and drug discovery. SVM models have black box character. Their predictions can be interpreted through feature weighting or the model-agnostic Shapley additive explanations (SHAP) formalism that locally approximates Shapley values (SVs) originating from game theory. We introduce an algorithm termed SV-expressed Tanimoto similarity (SVETA) for the exact calculation of SVs to explain SVM models employing the Tanimoto kernel, the gold standard for the assessment of molecular similarity. For a model system, the exact calculation of SVs is demonstrated. In an SVM-based compound classification task from drug discovery, only a limited correlation between exact SV and SHAP values is observed, prohibiting the use of approximate values for rationalizing predictions. For exemplary test compounds, atom-based mapping of prioritized features delineates coherent substructures that closely resemble those obtained by analyzing independently derived random forest models, thus providing consistent explanations.

•

SVETA: new methodology for explaining support vector machine (SVM) predictions

•

Tanimoto similarity-based SVM models are popular in chemistry

•

SVETA enables the calculation of exact Shapley values for rationalizing SVM models

•

SVETA-based feature mapping provides intuitive explanations of SVM decisions

In Silico Screening of Quorum Sensing Inhibitor Candidates Obtained by Chemical Similarity Search

Quorum sensing (QS) is a bacterial communication using signal molecules, by which they sense population density of their own species, leading to group behavior such as biofilm formation and virulence. Autoinducer-2 (AI2) is a QS signal molecule universally used by both gram-positive and gram-negative bacteria. Inhibition of QS mediated by AI2 is important for various practical applications, including prevention of gum-disease caused by biofilm formation of oral bacteria. In this research, molecular docking and molecular dynamics (MD) simulations were performed for molecules that are chemically similar to known AI2 inhibitors that might have a potential to be quorum sensing inhibitors. The molecules that form stable complexes with the AI2 receptor protein were found, suggesting that they could be developed as a novel AI2 inhibitors after further in vitro validation. The result suggests that combination of ligand-based drug design and computational methods such as MD simulation, and experimental verification, may lead to development of novel AI inhibitor, with a broad range of practical applications.

SMILES-based 2D-QSAR and similarity search for identification of potential new scaffolds for development of SARS-CoV-2 MPRO inhibitors

COVID-19, whose etiological agent is the SARS-CoV-2 virus, has caused over 537.5 million cases and killed over 6.3 million people since its discovery in 2019. Despite the recent development of the first drugs indicated for treating people already infected, the great need to develop new anti-SARS-CoV-2 drugs still exists, mainly due to the possible emergence of new variants of this virus and resistant strains of the current variants. Thus, this work presents the results of QSAR and similarity search studies based only on 2D structures from a set of 32 bicycloproline derivatives, aiming to quickly, reproducibly, and reliably identify potentially useful compounds as scaffolds of new major protease inhibitors (M^pro) of the virus. The obtained QSAR model is based only on topological molecular descriptors. The model has good internal and external statistics, is robust, and does not present a chance correlation. This model was used as one of the tools to support the virtual screening stage carried out in the SwissADME web tool. Five molecules, from an initial set of 2695 molecules, proved to be the most promising, as they were within the model’s applicability domain and linearity range, with low potential to cause carcinogenic, teratogenic, and reproductive toxicity effects and promising pharmacokinetic properties. These five compounds were then selected as the most competent to generate, in future studies, new anti-SARS-CoV-2 agents with drug-likeness properties suitable for use in therapy.

Creation of targeted compound libraries based on 3D shape recognition

In the emerging field of drug discovery, rapid virtual screening methods become extremely valuable, especially when dealing with ultra-large databases of organic small bioactive molecules. In this work, we present a fast, computationally resource-efficient, and simple workflow for screening targeted compound libraries generated from ultra-large virtual chemical space. This workflow aims to find compounds with similar molecular 3D shapes with reference ones, and at the same time to expand chemical diversity and to identify new and potentially active scaffolds. This pipeline ensures the enrichment of the generated libraries with novel chemotypes. Also, it was shown that delicate tailoring of the physicochemical parameters of the search set ensures that all library compounds will possess desired property distributions. A visual inspection has shown that found structures bind to the receptor in the same way as the reference ones. Using our screening workflow, we have created a number of conventional protein-targeted libraries: the GPCRs Targeted Library (531 K compounds) and the Protein Kinases Targeted Library (113 K compounds). The described pipeline and scripts are freely accessible at: https://github.com/ChemSpace-LLC/usrcat_sim .

Quantitative and Qualitative Analysis of the Anti-Proliferative Potential of the Pyrazole Scaffold in the Design of Anticancer Agents

The current work presents an objective overview of the impact of one important heterocyclic structure, the pyrazole ring, in the development of anti-proliferative drugs. A set of 1551 pyrazole derivatives were extracted from the National Cancer Institute (NCI) database, together with their growth inhibition effects (GI%) on the NCI’s panel of 60 cancer cell lines. The structures of these derivatives were analyzed based on the compounds’ averages of GI% values across NCI-60 cell lines and the averages of the values for the outlier cells. The distribution and the architecture of the Bemis–Murcko skeletons were analyzed, highlighting the impact of certain scaffold structures on the anti-proliferative effect’s potency and selectivity. The drug-likeness, chemical reactivity and promiscuity risks of the compounds were predicted using AMDETlab. The pyrazole ring proved to be a versatile scaffold for the design of anticancer drugs if properly substituted and if connected with other cyclic structures. The 1,3-diphenyl-pyrazole emerged as a useful scaffold for potent and targeted anticancer candidates.

Comparative Analyses of Medicinal Chemistry and Cheminformatics Filters with Accessible Implementation in Konstanz Information Miner (KNIME)

High-throughput virtual screening (HTVS) is, in conjunction with rapid advances in computer hardware, becoming a staple in drug design research campaigns and cheminformatics. In this context, virtual compound library design becomes crucial as it generally constitutes the first step where quality filtered databases are essential for the efficient downstream research. Therefore, multiple filters for compound library design were devised and reported in the scientific literature. We collected the most common filters in medicinal chemistry (PAINS, REOS, Aggregators, van de Waterbeemd, Oprea, Fichert, Ghose, Mozzicconacci, Muegge, Egan, Murcko, Veber, Ro3, Ro4, and Ro5) to facilitate their open access use and compared them. Then, we implemented these filters in the open platform Konstanz Information Miner (KNIME) as a freely accessible and simple workflow compatible with small or large compound databases for the benefit of the readers and for the help in the early drug design steps.

Indexing Ultrafast Shape-Based Descriptors in MongoDB to Identify TLR4 Pathway Agonists

A method is presented for an ultrafast shape-based search workflow for the screening of large compound collections, i.e., those of vendors. The three-dimensional shape of a molecule dictates its biological activity by enabling the molecule to fit into binding pockets of proteins. Quite often, distinctly different chemical compounds that have similar shapes can bind in a similar way. OpenEye pioneered an algorithm for comparing shapes of molecules by overlaying them in a computer and measuring differences between a query molecule and a target molecule. Overlaying shapes is a computationally intensive process and represents a bottleneck in searching for similar molecules. More recent publications describe alternative methods of overlaying molecules, which are accomplished by comparing shape-based descriptors. These methods were implemented in the Open Drug Discovery Toolkit (ODDT) package. We utilized a combination of open-source software packages like ODDT and RDkit to implement a workflow for ultrafast conformer generation and matching that does not require storing precomputed conformers on the file system or in memory. Moreover, the generated descriptors could be optionally stored in MongoDB for performing searches in the future. To speed up the search, we created a set of indexes from the transformed shape-based descriptors. We are in the process of calculating descriptors for multiple vendors, including Enamine's "REAL" collection of 1.2 billion compounds. Currently, the shape similarity search on more than 70 million compounds takes less than 8 s! We exemplified our methodology with the screen of compounds that can act as putative TLR4 agonists. The search was based on a literature-known small-molecule TLR4 agonist series. In due course, we identified compounds with novel structural motifs that were active in mouse and human TLR4 reporter cell lines.

Differentiating Inhibitors of Closely Related Protein Kinases with Single- or Multi-Target Activity via Explainable Machine Learning and Feature Analysis

Protein kinases are major drug targets. Most kinase inhibitors are directed against the adenosine triphosphate (ATP) cofactor binding site, which is largely conserved across the human kinome. Hence, such kinase inhibitors are often thought to be promiscuous. However, experimental evidence and activity data for publicly available kinase inhibitors indicate that this is not generally the case. We have investigated whether inhibitors of closely related human kinases with single- or multi-kinase activity can be differentiated on the basis of chemical structure. Therefore, a test system consisting of two distinct kinase triplets has been devised for which inhibitors with reported triple-kinase activities and corresponding single-kinase activities were assembled. Machine learning models derived on the basis of chemical structure distinguished between these multi- and single-kinase inhibitors with high accuracy. A model-independent explanatory approach was applied to identify structural features determining accurate predictions. For both kinase triplets, the analysis revealed decisive features contained in multi-kinase inhibitors. These features were found to be absent in corresponding single-kinase inhibitors, thus providing a rationale for successful machine learning. Mapping of features determining accurate predictions revealed that they formed coherent and chemically meaningful substructures that were characteristic of multi-kinase inhibitors compared with single-kinase inhibitors.