Quantifying the Hardness of Bioactivity Prediction Tasks for Transfer Learning.

Today, machine learning methods are widely employed in drug discovery. However, the chronic lack of data continues to hamper their further development, validation, and application. Several modern strategies aim to mitigate the challenges associated with data scarcity by learning from data on related tasks. These knowledge-sharing approaches encompass transfer learning, multitask learning, and meta-learning. A key question remaining to be answered for these approaches is about the extent to which their performance can benefit from the relatedness of available source (training) tasks; in other words, how difficult ("hard") a test task is to a model, given the available source tasks. This study introduces a new method for quantifying and predicting the hardness of a bioactivity prediction task based on its relation to the available training tasks. The approach involves the generation of protein and chemical representations and the calculation of distances between the bioactivity prediction task and the available training tasks. In the example of meta-learning on the FS-Mol data set, we demonstrate that the proposed task hardness metric is inversely correlated with performance (Pearson's correlation coefficient r = -0.72). The metric will be useful in estimating the task-specific gain in performance that can be achieved through meta-learning.

Active Learning Approach for Guiding Site-of-Metabolism Measurement and Annotation.

The ability to determine and predict metabolically labile atom positions in a molecule (also called "sites of metabolism" or "SoMs") is of high interest to the design and optimization of bioactive compounds, such as drugs, agrochemicals, and cosmetics. In recent years, several in silico models for SoM prediction have become available, many of which include a machine-learning component. The bottleneck in advancing these approaches is the coverage of distinct atom environments and rare and complex biotransformation events with high-quality experimental data. Pharmaceutical companies typically have measured metabolism data available for several hundred to several thousand compounds. However, even for metabolism experts, interpreting these data and assigning SoMs are challenging and time-consuming. Therefore, a significant proportion of the potential of the existing metabolism data, particularly in machine learning, remains dormant. Here, we report on the development and validation of an active learning approach that identifies the most informative atoms across molecular data sets for SoM annotation. The active learning approach, built on a highly efficient reimplementation of SoM predictor FAME 3, enables experts to prioritize their SoM experimental measurements and annotation efforts on the most rewarding atom environments. We show that this active learning approach yields competitive SoM predictors while requiring the annotation of only 20% of the atom positions required by FAME 3. The source code of the approach presented in this work is publicly available.

Identification of Natural Products Inhibiting SARS-CoV-2 by Targeting Viral Proteases: A Combined in Silico and in Vitro Approach.

In this study, an integrated in silico-in vitro approach was employed to discover natural products (NPs) active against SARS-CoV-2. The two SARS-CoV-2 viral proteases, i.e., main protease (Mpro) and papain-like protease (PLpro), were selected as targets for the in silico study. Virtual hits were obtained by docking more than 140,000 NPs and NP derivatives available in-house and from commercial sources, and 38 virtual hits were experimentally validated in vitro using two enzyme-based assays. Five inhibited the enzyme activity of SARS-CoV-2 Mpro by more than 60% at a concentration of 20 µM, and four of them with high potency (IC50 < 10 µM). These hit compounds were further evaluated for their antiviral activity against SARS-CoV-2 in Calu-3 cells. The results from the cell-based assay revealed three mulberry Diels-Alder-type adducts (MDAAs) from Morus alba with pronounced anti-SARS-CoV-2 activities. Sanggenons C (12), O (13), and G (15) showed IC50 values of 4.6, 8.0, and 7.6 µM and selectivity index values of 5.1, 3.1 and 6.5, respectively. The docking poses of MDAAs in SARS-CoV-2 Mpro proposed a butterfly-shaped binding conformation, which was supported by the results of saturation transfer difference NMR experiments and competitive 1H relaxation dispersion NMR spectroscopy.

Flindissone, a Limonoid Isolated from Trichilia prieuriana, Is an LXR Agonist.

In this study, the ability of six limonoids from Trichilia prieuriana (Meliaceae) to activate the liver X receptor (LXR) was assessed. One of these limonoids, flindissone, was shown to activate LXR by reporter-gene assays. Flindissone is a ring-intact limonoid, structurally similar to sterol-like LXR ligands. In endogenous cellular settings, flindissone showed an activity profile that is characteristic of LXR agonists. It induced cholesterol efflux in THP-1 macrophages by increasing the cholesterol transporter ABCA1 and ABCG1 gene expression. In HepG2 cells, flindissone induced the expression of IDOL, an LXR-target gene that is associated with the downregulation of the LDL receptor. However, unlike synthetic and similarly to sterol-based LXR agonists, flindissone did not induce the expression of the SREBP1c gene, a major transcription factor regulating de novo lipogenesis. Additionally, flindissone also appeared to be able to inhibit post-translational activation of SREBP1c. The results presented here reveal a natural product as a new LXR agonist and point to an additional property of T. prieuriana and other plant extracts containing flindissone.

Structural Insights into the Ligand-LsrK Kinase Binding Mode: A Step Forward in the Discovery of Novel Antimicrobial Agents.

LsrK is a bacterial kinase that triggers the quorum sensing, and it represents a druggable target for the identification of new agents for fighting antimicrobial resistance. Herein, we exploited tryptophan fluorescence spectroscopy (TFS) as a suitable technique for the identification of potential LsrK ligands from an in-house library of chemicals comprising synthetic compounds as well as secondary metabolites. Three secondary metabolites (Hib-ester, Hib-carbaldehyde and (R)-ASME) showed effective binding to LsrK, with KD values in the sub-micromolar range. The conformational changes were confirmed via circular dichroism and molecular docking results further validated the findings and displayed the specific mode of interaction. The activity of the identified compounds on the biofilm formation by some Staphylococcus spp. was investigated. Hib-carbaldehyde and (R)-ASME were able to reduce the production of biofilm, with (R)-ASME resulting in the most effective compound with an EC50 of 14 mg/well. The successful application of TFS highlights its usefulness in searching for promising LsrK inhibitor candidates with inhibitor efficacy against biofilm formation.

Ring systems in natural products: structural diversity, physicochemical properties, and coverage by synthetic compounds.

Covering: up to 2021The structural core of most small-molecule drugs is formed by a ring system, often derived from natural products. However, despite the importance of natural product ring systems in bioactive small molecules, there is still a lack of a comprehensive overview and understanding of natural product ring systems and how their full potential can be harnessed in drug discovery and related fields. Herein, we present a comprehensive cheminformatic analysis of the structural and physicochemical properties of 38 662 natural product ring systems, and the coverage of natural product ring systems by readily purchasable, synthetic compounds that are commonly explored in virtual screening and high-throughput screening. The analysis stands out by the use of comprehensive, curated data sets, the careful consideration of stereochemical information, and a robust analysis of the 3D molecular shape and electrostatic properties of ring systems. Among the key findings of this study are the facts that only about 2% of the ring systems observed in NPs are present in approved drugs but that approximately one in two NP ring systems are represented by ring systems with identical or related 3D shape and electrostatic properties in compounds that are typically used in (high-throughput) screening.

Validation strategies for target prediction methods.

Computational methods for target prediction, based on molecular similarity and network-based approaches, machine learning, docking and others, have evolved as valuable and powerful tools to aid the challenging task of mode of action identification for bioactive small molecules such as drugs and drug-like compounds. Critical to discerning the scope and limitations of a target prediction method is understanding how its performance was evaluated and reported. Ideally, large-scale prospective experiments are conducted to validate the performance of a model; however, this expensive and time-consuming endeavor is often not feasible. Therefore, to estimate the predictive power of a method, statistical validation based on retrospective knowledge is commonly used. There are multiple statistical validation techniques that vary in rigor. In this review we discuss the validation strategies employed, highlighting the usefulness and constraints of the validation schemes and metrics that are employed to measure and describe performance. We address the limitations of measuring only generalized performance, given that the underlying bioactivity and structural data are biased towards certain small-molecule scaffolds and target families, and suggest additional aspects of performance to consider in order to produce more detailed and realistic estimates of predictive power. Finally, we describe the validation strategies that were employed by some of the most thoroughly validated and accessible target prediction methods.

In Silico and In Vitro Approach to Assess Direct Allosteric AMPK Activators from Nature.

The 5'-adenosine monophosphate-activated protein kinase (AMPK) is an important metabolic regulator. Its allosteric drug and metabolite binding (ADaM) site was identified as an attractive target for direct AMPK activation and holds promise as a novel mechanism for the treatment of metabolic diseases. With the exception of lusianthridin and salicylic acid, no natural product (NP) is reported so far to directly target the ADaM site. For the streamlined assessment of direct AMPK activators from the pool of NPs, an integrated workflow using in silico and in vitro methods was applied. Virtual screening combining a 3D shape-based approach and docking identified 21 NPs and NP-like molecules that could potentially activate AMPK. The compounds were purchased and tested in an in vitro AMPK α 1 ß 1 γ 1 kinase assay. Two NP-like virtual hits were identified, which, at 30 µM concentration, caused a 1.65-fold (± 0.24) and a 1.58-fold (± 0.17) activation of AMPK, respectively. Intriguingly, using two different evaluation methods, we could not confirm the bioactivity of the supposed AMPK activator lusianthridin, which rebuts earlier reports.

Discovery of N-quinazolinone-4-hydroxy-2-quinolone-3-carboxamides as DNA gyrase B-targeted antibacterial agents.

Emerging drug resistance is generating an urgent need for novel and effective antibiotics. A promising target that has not yet been addressed by approved antibiotics is the bacterial DNA gyrase subunit B (GyrB), and GyrB inhibitors could be effective against drug-resistant bacteria, such as methicillin-resistant S. aureus (MRSA). Here, we used the 4-hydroxy-2-quinolone fragment to search the Specs database of purchasable compounds for potential inhibitors of GyrB and identified AG-690/11765367, or f1, as a novel and potent inhibitor of the target protein (IC50: 1.21 µM). Structural modification was used to further identify two more potent GyrB inhibitors: f4 (IC50: 0.31 µM) and f14 (IC50: 0.28 µM). Additional experiments indicated that compound f1 is more potent than the others in terms of antibacterial activity against MRSA (MICs: 4-8 µg/mL), non-toxic to HUVEC and HepG2 (CC50: approximately 50 µM), and metabolically stable (t1/2: > 372.8 min for plasma; 24.5 min for liver microsomes). In summary, this study showed that the discovered N-quinazolinone-4-hydroxy-2-quinolone-3-carboxamides are novel GyrB-targeted antibacterial agents; compound f1 is promising for further development.

Cryo-EM structure of pleconaril-resistant rhinovirus-B5 complexed to the antiviral OBR-5-340 reveals unexpected binding site.

Viral inhibitors, such as pleconaril and vapendavir, target conserved regions in the capsids of rhinoviruses (RVs) and enteroviruses (EVs) by binding to a hydrophobic pocket in viral capsid protein 1 (VP1). In resistant RVs and EVs, bulky residues in this pocket prevent their binding. However, recently developed pyrazolopyrimidines inhibit pleconaril-resistant RVs and EVs, and computational modeling has suggested that they also bind to the hydrophobic pocket in VP1. We studied the mechanism of inhibition of pleconaril-resistant RVs using RV-B5 (1 of the 7 naturally pleconaril-resistant rhinoviruses) and OBR-5-340, a bioavailable pyrazolopyrimidine with proven in vivo activity, and determined the 3D-structure of the protein-ligand complex to 3.6 Å with cryoelectron microscopy. Our data indicate that, similar to other capsid binders, OBR-5-340 induces thermostability and inhibits viral adsorption and uncoating. However, we found that OBR-5-340 attaches closer to the entrance of the pocket than most other capsid binders, whose viral complexes have been studied so far, showing only marginal overlaps of the attachment sites. Comparing the experimentally determined 3D structure with the control, RV-B5 incubated with solvent only and determined to 3.2 Å, revealed no gross conformational changes upon OBR-5-340 binding. The pocket of the naturally OBR-5-340-resistant RV-A89 likewise incubated with OBR-5-340 and solved to 2.9 Å was empty. Pyrazolopyrimidines have a rigid molecular scaffold and may thus be less affected by a loss of entropy upon binding. They interact with less-conserved regions than known capsid binders. Overall, pyrazolopyrimidines could be more suitable for the development of new, broadly active inhibitors.

Maximizing the Performance of Similarity-Based Virtual Screening Methods by Generating Synergy from the Integration of 2D and 3D Approaches.

Methods for the pairwise comparison of 2D and 3D molecular structures are established approaches in virtual screening. In this work, we explored three strategies for maximizing the virtual screening performance of these methods: (i) the merging of hit lists obtained from multi-compound screening using a single screening method, (ii) the merging of the hit lists obtained from 2D and 3D screening by parallel selection, and (iii) the combination of both of these strategies in an integrated approach. We found that any of these strategies led to a boost in virtual screening performance, with the clearest advantages observed for the integrated approach. On test sets for virtual screening, covering 50 pharmaceutically relevant proteins, the integrated approach, using sets of five query molecules, yielded, on average, an area under the receiver operating characteristic curve (AUC) of 0.84, an early enrichment among the top 1% of ranked compounds (EF1%) of 53.82 and a scaffold recovery rate among the top 1% of ranked compounds (SRR1%) of 0.50. In comparison, the 2D and 3D methods on their own (when using a single query molecule) yielded AUC values of 0.68 and 0.54, EF1% values of 19.96 and 17.52, and SRR1% values of 0.20 and 0.17, respectively. In conclusion, based on these results, the integration of 2D and 3D methods, via a (balanced) parallel selection strategy, is recommended, and, in particular, when combined with multi-query screening.

AKT1 and PTEN show the highest affinities among phosphoinositide binding proteins for the second messengers PtdIns(3,4,5)P3 and PtdIns(3,4)P2.

The phosphoinositides phosphatidylinositol-3,4,5-trisphosphate [PtdIns(3,4,5)P3] and phosphatidylinositol-3,4-bisphosphate [PtdIns(3,4)P2] function as second messengers and have been implicated in cancerogenesis. The signalling events downstream of PtdIns(3,4,5)P3 and PtdIns(3,4)P2 are mediated through a complex network of phosphoinositide binding effector proteins and phosphatases. In this study, we compared the phosphoinositide effector proteins AKT1, TAPP1, TAPP2, VAV1 and P-REX1 and the phosphoinositide phosphatases PTEN, SHIP1 and INPP4B for their binding affinities to PtdIns(3,4,5)P3 and/or PtdIns(3,4)P2 using Surface Plasmon Resonance. Our results demonstrate that all measured proteins except P-REX1 and VAV1 showed high affinity phosphoinositide binding with KD values in the nM to sub-nM range. Within the effector proteins, AKT1 showed the highest affinity for both PtdIns(3,4,5)P3 and PtdIns(3,4)P2. Of the phosphoinositide phosphatases PTEN displayed the highest affinity towards PtdIns(3,4,5)P3 and PtdIns(3,4)P2. The SHIP1 mutant E452K detected in carcinoma patients had a 100-fold increased affinity to PtdIns(3,4)P2 but not to PtdIns(3,4,5)P3 compared to SHIP1 WT. Distinct mutations in phosphoinositide binding proteins like the patient-derived SHIP1E452K mutant may be involved in the upregulation of PI(3,4)P2 -mediated signalling in tumor cells due to phosphoinositide trapping. Our results add further information to the complex hierarchy of phosphoinositide binding proteins helping to elucidate their functional role in cellular signal transduction.

NERDD: a web portal providing access to in silico tools for drug discovery.

SUMMARY: The New E-Resource for Drug Discovery (NERDD) is a quickly expanding web portal focused on the provision of peer-reviewed in silico tools for drug discovery. NERDD currently hosts tools for predicting the sites of metabolism (FAME) and metabolites (GLORY) of small organic molecules, for flagging compounds that are likely to interfere with biological assays (Hit Dexter), and for identifying natural products and natural product derivatives in large compound collections (NP-Scout). Several additional models and components are currently in development. AVAILABILITY AND IMPLEMENTATION: The NERDD web server is available at https://nerdd.zbh.uni-hamburg.de. Most tools are also available as software packages for local installation.

GLORYx: Prediction of the Metabolites Resulting from Phase 1 and Phase 2 Biotransformations of Xenobiotics.

Predicting the structures of metabolites formed in humans can provide advantageous insights for the development of drugs and other compounds. Here we present GLORYx, which integrates machine learning-based site of metabolism (SoM) prediction with reaction rule sets to predict and rank the structures of metabolites that could potentially be formed by phase 1 and/or phase 2 metabolism. GLORYx extends the approach from our previously developed tool GLORY, which predicted metabolite structures for cytochrome P450-mediated metabolism only. A robust approach to ranking the predicted metabolites is attained by using the SoM probabilities predicted by the FAME 3 machine learning models to score the predicted metabolites. On a manually curated test data set containing both phase 1 and phase 2 metabolites, GLORYx achieves a recall of 77% and an area under the receiver operating characteristic curve (AUC) of 0.79. Separate analysis of performance on a large amount of freely available phase 1 and phase 2 metabolite data indicates that achieving a meaningful ranking of predicted metabolites is more difficult for phase 2 than for phase 1 metabolites. GLORYx is freely available as a web server at https://nerdd.zbh.uni-hamburg.de/ and is also provided as a software package upon request. The data sets as well as all the reaction rules from this work are also made freely available.

In Silico Models to Predict the Perturbation of Molecular Initiating Events Related to Thyroid Hormone Homeostasis.

Disturbance of the thyroid hormone homeostasis has been associated with adverse health effects such as goiters and impaired mental development in humans and thyroid tumors in rats. In vitro and in silico methods for predicting the effects of small molecules on thyroid hormone homeostasis are currently being explored as alternatives to animal experiments, but are still in an early stage of development. The aim of this work was the development of a battery of in silico models for a set of targets involved in molecular initiating events of thyroid hormone homeostasis: deiodinases 1, 2, and 3, thyroid peroxidase (TPO), thyroid hormone receptor (TR), sodium/iodide symporter, thyrotropin-releasing hormone receptor, and thyroid-stimulating hormone receptor. The training data sets were compiled from the ToxCast database and related scientific literature. Classical statistical approaches as well as several machine learning methods (including random forest, support vector machine, and neural networks) were explored in combination with three data balancing techniques. The models were trained on molecular descriptors and fingerprints and evaluated on holdout data. Furthermore, multi-task neural networks combining several end points were investigated as a possible way to improve the performance of models for which the experimental data available for model training are limited. Classifiers for TPO and TR performed particularly well, with F1 scores of 0.83 and 0.81 on the holdout data set, respectively. Models for the other studied targets yielded F1 scores of up to 0.77. An in-depth analysis of the reliability of predictions was performed for the most relevant models. All data sets used in this work for model development and validation are available in the Supporting Information.

Skin Doctor CP: Conformal Prediction of the Skin Sensitization Potential of Small Organic Molecules.

Skin sensitization potential or potency is an important end point in the safety assessment of new chemicals and new chemical mixtures. Formerly, animal experiments such as the local lymph node assay (LLNA) were the main form of assessment. Today, however, the focus lies on the development of nonanimal testing approaches (i.e., in vitro and in chemico assays) and computational models. In this work, we investigate, based on publicly available LLNA data, the ability of aggregated, Mondrian conformal prediction classifiers to differentiate between non- sensitizing and sensitizing compounds as well as between two levels of skin sensitization potential (weak to moderate sensitizers, and strong to extreme sensitizers). The advantage of the conformal prediction framework over other modeling approaches is that it assigns compounds to activity classes only if a defined minimum level of confidence is reached for the individual predictions. This eliminates the need for applicability domain criteria that often are arbitrary in their nature and less flexible. Our new binary classifier, named Skin Doctor CP, differentiates nonsensitizers from sensitizers with a higher reliability-to-efficiency ratio than the corresponding nonconformal prediction workflow that we presented earlier. When tested on a set of 257 compounds at the significance levels of 0.10 and 0.30, the model reached an efficiency of 0.49 and 0.92, and an accuracy of 0.83 and 0.75, respectively. In addition, we developed a ternary classification workflow to differentiate nonsensitizers, weak to moderate sensitizers, and strong to extreme sensitizers. Although this model achieved satisfactory overall performance (accuracies of 0.90 and 0.73, and efficiencies of 0.42 and 0.90, at significance levels 0.10 and 0.30, respectively), it did not obtain satisfying class-wise results (at a significance level of 0.30, the validities obtained for nonsensitizers, weak to moderate sensitizers, and strong to extreme sensitizers were 0.70, 0.58, and 0.63, respectively). We argue that the model is, in consequence, unable to reliably identify strong to extreme sensitizers and suggest that other ternary models derived from the currently accessible LLNA data might suffer from the same problem. Skin Doctor CP is available via a public web service at https://nerdd.zbh.uni-hamburg.de/skinDoctorII/.

ChemBioSim: Enhancing Conformal Prediction of In Vivo Toxicity by Use of Predicted Bioactivities.

Computational methods such as machine learning approaches have a strong track record of success in predicting the outcomes of in vitro assays. In contrast, their ability to predict in vivo endpoints is more limited due to the high number of parameters and processes that may influence the outcome. Recent studies have shown that the combination of chemical and biological data can yield better models for in vivo endpoints. The ChemBioSim approach presented in this work aims to enhance the performance of conformal prediction models for in vivo endpoints by combining chemical information with (predicted) bioactivity assay outcomes. Three in vivo toxicological endpoints, capturing genotoxic (MNT), hepatic (DILI), and cardiological (DICC) issues, were selected for this study due to their high relevance for the registration and authorization of new compounds. Since the sparsity of available biological assay data is challenging for predictive modeling, predicted bioactivity descriptors were introduced instead. Thus, a machine learning model for each of the 373 collected biological assays was trained and applied on the compounds of the in vivo toxicity data sets. Besides the chemical descriptors (molecular fingerprints and physicochemical properties), these predicted bioactivities served as descriptors for the models of the three in vivo endpoints. For this study, a workflow based on a conformal prediction framework (a method for confidence estimation) built on random forest models was developed. Furthermore, the most relevant chemical and bioactivity descriptors for each in vivo endpoint were preselected with lasso models. The incorporation of bioactivity descriptors increased the mean F1 scores of the MNT model from 0.61 to 0.70 and for the DICC model from 0.72 to 0.82 while the mean efficiencies increased by roughly 0.10 for both endpoints. In contrast, for the DILI endpoint, no significant improvement in model performance was observed. Besides pure performance improvements, an analysis of the most important bioactivity features allowed detection of novel and less intuitive relationships between the predicted biological assay outcomes used as descriptors and the in vivo endpoints. This study presents how the prediction of in vivo toxicity endpoints can be improved by the incorporation of biological information-which is not necessarily captured by chemical descriptors-in an automated workflow without the need for adding experimental workload for the generation of bioactivity descriptors as predicted outcomes of bioactivity assays were utilized. All bioactivity CP models for deriving the predicted bioactivities, as well as the in vivo toxicity CP models, can be freely downloaded from https://doi.org/10.5281/zenodo.4761225.

CYPlebrity: Machine learning models for the prediction of inhibitors of cytochrome P450 enzymes.

The vast majority of approved drugs are metabolized by the five major cytochrome P450 (CYP) isozymes, 1A2, 2C9, 2C19, 2D6 and 3A4. Inhibition of CYP isozymes can cause drug-drug interactions with severe pharmacological and toxicological consequences. Computational methods for the fast and reliable prediction of the inhibition of CYP isozymes by small molecules are therefore of high interest and relevance to pharmaceutical companies and a host of other industries, including the cosmetics and agrochemical industries. Today, a large number of machine learning models for predicting the inhibition of the major CYP isozymes by small molecules are available. With this work we aim to go beyond the coverage of existing models, by combining data from several major public and proprietary sources. More specifically, we used up to 18815 compounds with measured bioactivities to train random forest classification models for the individual CYP isozymes. A major advantage of the new data collection over existing ones is the better representation of the minority class, the CYP inhibitors. With the new data collection we achieved inhibitor-to-non-inhibitor ratios in the order of 1:1 (CYP1A2) to 1:3 (CYP2D6). We show that our models reach competitive performance on external data, with Matthews correlation coefficients (MCCs) ranging from 0.62 (CYP2C19) to 0.70 (CYP2D6), and areas under the receiver operating characteristic curve (AUCs) between 0.89 (CYP2C19) and 0.92 (CYPs 2D6 and 3A4). Importantly, the models show a high level of robustness, reflected in a good predictivity also for compounds that are structurally dissimilar to the compounds represented in the training data. The best models presented in this work are freely accessible for academic research via a web service.

Antibacterial activity of xylose-derived LpxC inhibitors - Synthesis, biological evaluation and molecular docking studies.

LpxC inhibitors represent a promising class of novel antibiotics selectively combating Gram-negative bacteria. In chiral pool syntheses starting from D- and L-xylose, a series of four 2r,3c,4t-configured C-furanosidic LpxC inhibitors was obtained. The synthesized hydroxamic acids were tested for antibacterial and LpxC inhibitory activity, the acquired biological data were compared with those of previously synthesized C-furanosides, and molecular docking studies were performed to rationalize the observed structure-activity relationships. Additionally, bacterial uptake and susceptibility to efflux pump systems were investigated for the most promising stereoisomers.

BonMOLière: Small-Sized Libraries of Readily Purchasable Compounds, Optimized to Produce Genuine Hits in Biological Screens across the Protein Space.

Experimental screening of large sets of compounds against macromolecular targets is a key strategy to identify novel bioactivities. However, large-scale screening requires substantial experimental resources and is time-consuming and challenging. Therefore, small to medium-sized compound libraries with a high chance of producing genuine hits on an arbitrary protein of interest would be of great value to fields related to early drug discovery, in particular biochemical and cell research. Here, we present a computational approach that incorporates drug-likeness, predicted bioactivities, biological space coverage, and target novelty, to generate optimized compound libraries with maximized chances of producing genuine hits for a wide range of proteins. The computational approach evaluates drug-likeness with a set of established rules, predicts bioactivities with a validated, similarity-based approach, and optimizes the composition of small sets of compounds towards maximum target coverage and novelty. We found that, in comparison to the random selection of compounds for a library, our approach generates substantially improved compound sets. Quantified as the "fitness" of compound libraries, the calculated improvements ranged from +60% (for a library of 15,000 compounds) to +184% (for a library of 1000 compounds). The best of the optimized compound libraries prepared in this work are available for download as a dataset bundle ("BonMOLière").