Using information from historical high-throughput screens to predict active compounds

Cell Painting-based bioactivity prediction boosts high-throughput screening hit-rates and compound diversity

Identifying active compounds for a target is a time- and resource-intensive task in early drug discovery. Accurate bioactivity prediction using morphological profiles could streamline the process, enabling smaller, more focused compound screens. We investigate the potential of deep learning on unrefined single-concentration activity readouts and Cell Painting data, to predict compound activity across 140 diverse assays. We observe an average ROC-AUC of 0.744 ± 0.108 with 62% of assays achieving ≥0.7, 30% ≥0.8, and 7% ≥0.9. In many cases, the high prediction performance can be achieved using only brightfield images instead of multichannel fluorescence images. A comprehensive analysis shows that Cell Painting-based bioactivity prediction is robust across assay types, technologies, and target classes, with cell-based assays and kinase targets being particularly well-suited for prediction. Experimental validation confirms the enrichment of active compounds. Our findings indicate that models trained on Cell Painting data, combined with a small set of single-concentration data points, can reliably predict the activity of a compound library across diverse targets and assays while maintaining high hit rates and scaffold diversity. This approach has the potential to reduce the size of screening campaigns, saving time and resources, and enabling primary screening with more complex assays.

Enhancing the Small-Scale Screenable Biological Space beyond Known Chemogenomics Libraries with Gray Chemical Matter─Compounds with Novel Mechanisms from High-Throughput Screening Profiles

Phenotypic assays have become an established approach to drug discovery. Greater disease relevance is often achieved through cellular models with increased complexity and more detailed readouts, such as gene expression or advanced imaging. However, the intricate nature and cost of these assays impose limitations on their screening capacity, often restricting screens to well-characterized small compound sets such as chemogenomics libraries. Here, we outline a cheminformatics approach to identify a small set of compounds with likely novel mechanisms of action (MoAs), expanding the MoA search space for throughput limited phenotypic assays. Our approach is based on mining existing large-scale, phenotypic high-throughput screening (HTS) data. It enables the identification of chemotypes that exhibit selectivity across multiple cell-based assays, which are characterized by persistent and broad structure activity relationships (SAR). We validate the effectiveness of our approach in broad cellular profiling assays (Cell Painting, DRUG-seq, and Promotor Signature Profiling) and chemical proteomics experiments. These experiments revealed that the compounds behave similarly to known chemogenetic libraries, but with a notable bias toward novel protein targets. To foster collaboration and advance research in this area, we have curated a public set of such compounds based on the PubChem BioAssay dataset and made it available for use by the scientific community.

Applications of Machine Learning (ML) and Mathematical Modeling (MM) in Healthcare with Special Focus on Cancer Prognosis and Anticancer Therapy: Current Status and Challenges

The use of data-driven high-throughput analytical techniques, which has given rise to computational oncology, is undisputed. The widespread use of machine learning (ML) and mathematical modeling (MM)-based techniques is widely acknowledged. These two approaches have fueled the advancement in cancer research and eventually led to the uptake of telemedicine in cancer care. For diagnostic, prognostic, and treatment purposes concerning different types of cancer research, vast databases of varied information with manifold dimensions are required, and indeed, all this information can only be managed by an automated system developed utilizing ML and MM. In addition, MM is being used to probe the relationship between the pharmacokinetics and pharmacodynamics (PK/PD interactions) of anti-cancer substances to improve cancer treatment, and also to refine the quality of existing treatment models by being incorporated at all steps of research and development related to cancer and in routine patient care. This review will serve as a consolidation of the advancement and benefits of ML and MM techniques with a special focus on the area of cancer prognosis and anticancer therapy, leading to the identification of challenges (data quantity, ethical consideration, and data privacy) which are yet to be fully addressed in current studies.

Discovery of hematopoietic progenitor kinase 1 inhibitors using machine learning-based screening and free energy perturbation

Hematopoietic progenitor kinase 1 (HPK1) is a key negative regulator of T-cell receptor (TCR) signaling and a promising target for cancer immunotherapy. The development of novel HPK1 inhibitors is challenging yet promising. In this study, we used a combination of machine learning (ML)-based virtual screening and free energy perturbation (FEP) calculations to identify novel HPK1 inhibitors. ML-based screening yielded 10 potent HPK1 inhibitors (IC50 < 1 μM). The FEP-guided modification of the in-house false-positive hit, DW21302, revealed that a single key atom change could trigger activity cliffs. The resulting DW21302-A was a potent HPK1 inhibitor (IC50 = 2.1 nM) and potently inhibited cellular HPK1 signaling and enhanced T-cell function. Molecular dynamics (MD) simulations and ADME predictions confirmed DW21302-A as candidate compound. This study provides new strategies and chemical scaffolds for HPK1 inhibitor development.Communicated by Ramaswamy H. Sarma.

Deep Learning with Geometry-Enhanced Molecular Representation for Augmentation of Large-Scale Docking-Based Virtual Screening

Structure-based virtual screening has been a crucial tool in drug discovery for decades. However, as the chemical space expands, the existing structure-based virtual screening techniques based on molecular docking and scoring struggle to handle billion-entry ultralarge libraries due to the high computational cost. To address this challenge, people have resorted to machine learning techniques to enhance structure-based virtual screening for efficiently exploring the vast chemical space. In those cases, compounds are usually treated as sequential strings or two-dimensional topology graphs, limiting their ability to incorporate three-dimensional structural information for downstream tasks. We herein propose a novel deep learning protocol, GEM-Screen, which utilizes the geometry-enhanced molecular representation of the compounds docking to a specific target and is trained on docking scores of a small fraction of a library through an active learning strategy to approximate the docking outcome for yet nontraining entries. This protocol is applied to virtual screening campaigns against the AmpC and D4 targets, demonstrating that GEM-Screen enriches more than 90% of the hit scaffolds for AmpC in the top 4% of model predictions and more than 80% of the hit scaffolds for D4 in the same top-ranking size of library. GEM-Screen can be used in conjunction with traditional docking programs for docking of only the top-ranked compounds to avoid the exhaustive docking of the whole library, thus allowing for discovering top-scoring compounds from billion-entry libraries in a rapid yet accurate fashion.

Overfit deep neural network for predicting drug-target interactions

Drug-target interactions (DTIs) prediction is an important step in drug discovery. As traditional biological experiments or high-throughput screening are high cost and time-consuming, many deep learning models have been developed. Overfitting must be avoided when training deep learning models. We propose a simple framework, called OverfitDTI, for DTI prediction. In OverfitDTI, a deep neural network (DNN) model is overfit to sufficiently learn the features of the chemical space of drugs and the biological space of targets. The weights of trained DNN model form an implicit representation of the nonlinear relationship between drugs and targets. Performance of OverfitDTI on three public datasets showed that the overfit DNN models fit the nonlinear relationship with high accuracy. We identified fifteen compounds that interacted with TEK, a receptor tyrosine kinase contributing to vascular homeostasis, and the predicted AT9283 and dorsomorphin were experimentally demonstrated as inhibitors of TEK in human umbilical vein endothelial cells (HUVECs).

High-Throughput Synthesis and Machine Learning Assisted Design of Photodegradable Hydrogels

Due to the large chemical space, the design of functional and responsive soft materials poses many challenges but also offers a wide range of opportunities in terms of the scope of possible properties. Herein, an experimental workflow for miniaturized combinatorial high-throughput screening of functional hydrogel libraries is reported. The data created from the analysis of the photodegradation process of more than 900 different types of hydrogel pads are used to train a machine learning model for automated decision making. Through iterative model optimization based on Bayesian optimization, a substantial improvement in response properties is achieved and thus expanded the scope of material properties obtainable within the chemical space of hydrogels in the study. It is therefore demonstrated that the potential of combining miniaturized high-throughput experiments with smart optimization algorithms for cost and time efficient optimization of materials properties.

Combining metabolome and clinical indicators with machine learning provides some promising diagnostic markers to precisely detect smear-positive/negative pulmonary tuberculosis

Background

Tuberculosis (TB) had been the leading lethal infectious disease worldwide for a long time (2014–2019) until the COVID-19 global pandemic, and it is still one of the top 10 death causes worldwide. One important reason why there are so many TB patients and death cases in the world is because of the difficulties in precise diagnosis of TB using common detection methods, especially for some smear-negative pulmonary tuberculosis (SNPT) cases. The rapid development of metabolome and machine learning offers a great opportunity for precision diagnosis of TB. However, the metabolite biomarkers for the precision diagnosis of smear-positive and smear-negative pulmonary tuberculosis (SPPT/SNPT) remain to be uncovered. In this study, we combined metabolomics and clinical indicators with machine learning to screen out newly diagnostic biomarkers for the precise identification of SPPT and SNPT patients.

Methods

Untargeted plasma metabolomic profiling was performed for 27 SPPT patients, 37 SNPT patients and controls. The orthogonal partial least squares-discriminant analysis (OPLS-DA) was then conducted to screen differential metabolites among the three groups. Metabolite enriched pathways, random forest (RF), support vector machines (SVM) and multilayer perceptron neural network (MLP) were performed using Metaboanalyst 5.0, “caret” R package, “e1071” R package and “Tensorflow” Python package, respectively.

Results

Metabolomic analysis revealed significant enrichment of fatty acid and amino acid metabolites in the plasma of SPPT and SNPT patients, where SPPT samples showed a more serious dysfunction in fatty acid and amino acid metabolisms. Further RF analysis revealed four optimized diagnostic biomarker combinations including ten features (two lipid/lipid-like molecules and seven organic acids/derivatives, and one clinical indicator) for the identification of SPPT, SNPT patients and controls with high accuracy (83–93%), which were further verified by SVM and MLP. Among them, MLP displayed the best classification performance on simultaneously precise identification of the three groups (94.74%), suggesting the advantage of MLP over RF/SVM to some extent.

Conclusions

Our findings reveal plasma metabolomic characteristics of SPPT and SNPT patients, provide some novel promising diagnostic markers for precision diagnosis of various types of TB, and show the potential of machine learning in screening out biomarkers from big data.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-022-07694-8.

Pharmacological affinity fingerprints derived from bioactivity data for the identification of designer drugs

Facing the continuous emergence of new psychoactive substances (NPS) and their threat to public health, more effective methods for NPS prediction and identification are critical. In this study, the pharmacological affinity fingerprints (Ph-fp) of NPS compounds were predicted by Random Forest classification models using bioactivity data from the ChEMBL database. The binary Ph-fp is the vector consisting of a compound’s activity against a list of molecular targets reported to be responsible for the pharmacological effects of NPS. Their performance in similarity searching and unsupervised clustering was assessed and compared to 2D structure fingerprints Morgan and MACCS (1024-bits ECFP4 and 166-bits SMARTS-based MACCS implementation of RDKit). The performance in retrieving compounds according to their pharmacological categorizations is influenced by the predicted active assay counts in Ph-fp and the choice of similarity metric. Overall, the comparative unsupervised clustering analysis suggests the use of a classification model with Morgan fingerprints as input for the construction of Ph-fp. This combination gives satisfactory clustering performance based on external and internal clustering validation indices.

Wet-dry-wet drug screen leads to the synthesis of TS1, a novel compound reversing lung fibrosis through inhibition of myofibroblast differentiation

Therapies halting the progression of fibrosis are ineffective and limited. Activated myofibroblasts are emerging as important targets in the progression of fibrotic diseases. Previously, we performed a high-throughput screen on lung fibroblasts and subsequently demonstrated that the inhibition of myofibroblast activation is able to prevent lung fibrosis in bleomycin-treated mice. High-throughput screens are an ideal method of repurposing drugs, yet they contain an intrinsic limitation, which is the size of the library itself. Here, we exploited the data from our "wet" screen and used "dry" machine learning analysis to virtually screen millions of compounds, identifying novel anti-fibrotic hits which target myofibroblast differentiation, many of which were structurally related to dopamine. We synthesized and validated several compounds ex vivo ("wet") and confirmed that both dopamine and its derivative TS1 are powerful inhibitors of myofibroblast activation. We further used RNAi-mediated knock-down and demonstrated that both molecules act through the dopamine receptor 3 and exert their anti-fibrotic effect by inhibiting the canonical transforming growth factor β pathway. Furthermore, molecular modelling confirmed the capability of TS1 to bind both human and mouse dopamine receptor 3. The anti-fibrotic effect on human cells was confirmed using primary fibroblasts from idiopathic pulmonary fibrosis patients. Finally, TS1 prevented and reversed disease progression in a murine model of lung fibrosis. Both our interdisciplinary approach and our novel compound TS1 are promising tools for understanding and combating lung fibrosis.

Recent applications of quantitative systems pharmacology and machine learning models across diseases

Quantitative systems pharmacology (QSP) is a quantitative and mechanistic platform describing the phenotypic interaction between drugs, biological networks, and disease conditions to predict optimal therapeutic response. In this meta-analysis study, we review the utility of the QSP platform in drug development and therapeutic strategies based on recent publications (2019-2021). We gathered recent original QSP models and described the diversity of their applications based on therapeutic areas, methodologies, software platforms, and functionalities. The collection and investigation of these publications can assist in providing a repository of recent QSP studies to facilitate the discovery and further reusability of QSP models. Our review shows that the largest number of QSP efforts in recent years is in Immuno-Oncology. We also addressed the benefits of integrative approaches in this field by presenting the applications of Machine Learning methods for drug discovery and QSP models. Based on this meta-analysis, we discuss the advantages and limitations of QSP models and propose fields where the QSP approach constitutes a valuable interface for more investigations to tackle complex diseases and improve drug development.

Predicting the Skin Sensitization Potential of Small Molecules with Machine Learning Models Trained on Biologically Meaningful Descriptors

In recent years, a number of machine learning models for the prediction of the skin sensitization potential of small organic molecules have been reported and become available. These models generally perform well within their applicability domains but, as a result of the use of molecular fingerprints and other non-intuitive descriptors, the interpretability of the existing models is limited. The aim of this work is to develop a strategy to replace the non-intuitive features by predicted outcomes of bioassays. We show that such replacement is indeed possible and that as few as ten interpretable, predicted bioactivities are sufficient to reach competitive performance. On a holdout data set of 257 compounds, the best model (“Skin Doctor CP:Bio”) obtained an efficiency of 0.82 and an MCC of 0.52 (at the significance level of 0.20). Skin Doctor CP:Bio is available free of charge for academic research. The modeling strategies explored in this work are easily transferable and could be adopted for the development of more interpretable machine learning models for the prediction of the bioactivity and toxicity of small organic compounds.

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.

GHOST: Adjusting the Decision Threshold to Handle Imbalanced Data in Machine Learning

Machine learning classifiers trained on class imbalanced data are prone to overpredict the majority class. This leads to a larger misclassification rate for the minority class, which in many real-world applications is the class of interest. For binary data, the classification threshold is set by default to 0.5 which, however, is often not ideal for imbalanced data. Adjusting the decision threshold is a good strategy to deal with the class imbalance problem. In this work, we present two different automated procedures for the selection of the optimal decision threshold for imbalanced classification. A major advantage of our procedures is that they do not require retraining of the machine learning models or resampling of the training data. The first approach is specific for random forest (RF), while the second approach, named GHOST, can be potentially applied to any machine learning classifier. We tested these procedures on 138 public drug discovery data sets containing structure-activity data for a variety of pharmaceutical targets. We show that both thresholding methods improve significantly the performance of RF. We tested the use of GHOST with four different classifiers in combination with two molecular descriptors, and we found that most classifiers benefit from threshold optimization. GHOST also outperformed other strategies, including random undersampling and conformal prediction. Finally, we show that our thresholding procedures can be effectively applied to real-world drug discovery projects, where the imbalance and characteristics of the data vary greatly between the training and test sets.

Biological activity-based modeling identifies antiviral leads against SARS-CoV-2

Computational approaches for drug discovery, such as quantitative structure-activity relationship, rely on structural similarities of small molecules to infer biological activity but are often limited to identifying new drug candidates in the chemical spaces close to known ligands. Here we report a biological activity-based modeling (BABM) approach, in which compound activity profiles established across multiple assays are used as signatures to predict compound activity in other assays or against a new target. This approach was validated by identifying candidate antivirals for Zika and Ebola viruses based on high-throughput screening data. BABM models were then applied to predict 311 compounds with potential activity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Of the predicted compounds, 32% had antiviral activity in a cell culture live virus assay, the most potent compounds showing a half-maximal inhibitory concentration in the nanomolar range. Most of the confirmed anti-SARS-CoV-2 compounds were found to be viral entry inhibitors and/or autophagy modulators. The confirmed compounds have the potential to be further developed into anti-SARS-CoV-2 therapies.

Discovery of Novel eEF2K Inhibitors Using HTS Fingerprint Generated from Predicted Profiling of Compound-Protein Interactions

Background: Eukaryotic elongation factor 2 kinase (eEF2K) regulates the elongation stage of protein synthesis by phosphorylating eEF2, a process related to various diseases including cancer and cardiovascular and neurodegenerative diseases. In this study, we describe the identification of novel eEF2K inhibitors using high-throughput screening fingerprints (HTSFP) generated from predicted profiling of compound-protein interactions (CPIs). Methods: We utilized computationally generated HTSFPs referred to as chemical genomics-based fingerprint (CGBFP). Generally, HTSFPs are generated from multiple biochemical or cell-based assay data. On the other hand, CGBFPs are generated from computational prediction of CPIs using the Chemical Genomics-Based Virtual Screening (CGBVS) method. Therefore, CGBFPs do not have missing information mainly caused by the absence of assay data. Results: Chemogenomics-Based Similarity Profiling (CGBSP) of the screening library (2.6 million compounds) yielded 27 compounds which were evaluated for in vitro eEF2K inhibitory activity. Three compounds with interesting results were identified. Compounds 2 (IC50 = 11.05 μM) and 4 (IC50 = 43.54 μM) are thieno[2,3-b]pyridine derivatives that have the same scaffolds with a known eEF2K inhibitor, while compound 13 (IC50 = 70.13 μM) was a new thiophene-2-amine-type eEF2K inhibitor. Conclusions: CGBSP supplied an efficient strategy in the identification of novel eEF2K inhibitors and provided useful scaffolds for optimization.

Role of artificial intelligence in peptide vaccine design against RNA viruses

RNA viruses have high rate of replication and mutation that help them adapt and change according to their environmental conditions. Many viral mutants are the cause of various severe and lethal diseases. Vaccines, on the other hand have the capacity to protect us from infectious diseases by eliciting antibody or cell-mediated immune responses that are pathogen-specific. While there are a few reviews pertaining to the use of artificial intelligence (AI) for SARS-COV-2 vaccine development, none focus on peptide vaccination for RNA viruses and the important role played by AI in it. Peptide vaccine which is slowly coming to be recognized as a safe and effective vaccination strategy has the capacity to overcome the mutant escape problem which is also being currently faced by SARS-COV-2 vaccines in circulation.Here we review the present scenario of peptide vaccines which are developed using mathematical and computational statistics methods to prevent the spread of disease caused by RNA viruses. We also focus on the importance and current stage of AI and mathematical evolutionary modeling using machine learning tools in the establishment of these new peptide vaccines for the control of viral disease.

Prediction Machines: Applied Machine Learning for Therapeutic Protein Design and Development

The rapid growth in technological advances and quantity of scientific data over the past decade has led to several challenges including data storage and analysis. Accurate models of complex datasets were previously difficult to develop and interpret. However, improvements in machine learning algorithms have since enabled unparalleled classification and prediction capabilities. The application of machine learning can be seen throughout diverse industries due to their ease of use and interpretability. In this review, we describe popular machine learning algorithms and highlight their application in pharmaceutical protein development. Machine learning models have now been applied to better understand the nonlinear concentration dependent viscosity of protein solutions, predict protein oxidation and deamidation rates, classify sub-visible particles and compare the physical stability of proteins. We also applied several machine learning algorithms using previously published data and describe models with improved predictions and classification. The authors hope that this review can be used as a resource to others and encourage continued application of machine learning algorithms to problems in pharmaceutical protein development.

Early lung cancer diagnostic biomarker discovery by machine learning methods

•

Early diagnosis could improve lung cancer survival rate.

•

The availability of blood-based screening could increase lung cancer patient uptake.

•

An interdisciplinary mechanism combines metabolomics and machine learning methods.

•

Metabolic biomarkers could be potential screening biomarkers for early detection of lung cancer.

•

Naïve Bayes is recommended as an exploitable tool for early lung tumor prediction.

Early diagnosis has been proved to improve survival rate of lung cancer patients. The availability of blood-based screening could increase early lung cancer patient uptake. Our present study attempted to discover Chinese patients’ plasma metabolites as diagnostic biomarkers for lung cancer. In this work, we use a pioneering interdisciplinary mechanism, which is firstly applied to lung cancer, to detect early lung cancer diagnostic biomarkers by combining metabolomics and machine learning methods. We collected total 110 lung cancer patients and 43 healthy individuals in our study. Levels of 61 plasma metabolites were from targeted metabolomic study using LC-MS/MS. A specific combination of six metabolic biomarkers note-worthily enabling the discrimination between stage I lung cancer patients and healthy individuals (AUC = 0.989, Sensitivity = 98.1%, Specificity = 100.0%). And the top 5 relative importance metabolic biomarkers developed by FCBF algorithm also could be potential screening biomarkers for early detection of lung cancer. Naïve Bayes is recommended as an exploitable tool for early lung tumor prediction. This research will provide strong support for the feasibility of blood-based screening, and bring a more accurate, quick and integrated application tool for early lung cancer diagnostic. The proposed interdisciplinary method could be adapted to other cancer beyond lung cancer.

Harnessing In Silico, In Vitro, and In Vivo Data to Understand the Toxicity Landscape of Polycyclic Aromatic Compounds (PACs)

Polycyclic aromatic compounds (PACs) are compounds with a minimum of two six-atom aromatic fused rings. PACs arise from incomplete combustion or thermal decomposition of organic matter and are ubiquitous in the environment. Within PACs, carcinogenicity is generally regarded to be the most important public health concern. However, toxicity in other systems (reproductive and developmental toxicity, immunotoxicity) has also been reported. Despite the large number of PACs identified in the environment, research attention to understand exposure and health effects of PACs has focused on a relatively limited subset, namely polycyclic aromatic hydrocarbons (PAHs), the PACs with only carbon and hydrogen atoms. To triage the rest of the vast number of PACs for more resource-intensive testing, we developed a data-driven approach to contextualize hazard characterization of PACs, by leveraging the available data from various data streams (in silico toxicity, in vitro activity, structural fingerprints, and in vivo data availability). The PACs were clustered on the basis of their in silico toxicity profiles containing predictions from 8 different categories (carcinogenicity, cardiotoxicity, developmental toxicity, genotoxicity, hepatotoxicity, neurotoxicity, reproductive toxicity, and urinary toxicity). We found that PACs with the same parent structure (e.g., fluorene) could have diverse in silico toxicity profiles. In contrast, PACs with similar substituted groups (e.g., alkylated-PAHs) or heterocyclics (e.g., N-PACs) with varying ring sizes could have similar in silico toxicity profiles, suggesting that these groups are better candidates for toxicity read-across analysis. The clusters/regions associated with certain in silico toxicity, in vitro activity, and structural fingerprints were identified. We found that genotoxicity/carcinogenicity (in silico toxicity) and xenobiotic homeostasis and stress response (in vitro activity), respectively, dominate the toxicity/activity variation seen in the PACs. The "hot spots" with enriched toxicity/activity in conjunction with availability of in vivo carcinogenicity data revealed regions of either data-poor (hydroxylated-PAHs) or data-rich (unsubstituted, parent PAHs) PACs. These regions offer potential targets for prioritization of further in vivo assessment and for chemical read-across efforts. The analysis results are searchable through an interactive web application (https://ntp.niehs.nih.gov/go/pacs_tableau), allowing for alternative hypothesis generation.

Machine learning and AI-based approaches for bioactive ligand discovery and GPCR-ligand recognition

In the last decade, machine learning and artificial intelligence applications have received a significant boost in performance and attention in both academic research and industry. The success behind most of the recent state-of-the-art methods can be attributed to the latest developments in deep learning. When applied to various scientific domains that are concerned with the processing of non-tabular data, for example, image or text, deep learning has been shown to outperform not only conventional machine learning but also highly specialized tools developed by domain experts. This review aims to summarize AI-based research for GPCR bioactive ligand discovery with a particular focus on the most recent achievements and research trends. To make this article accessible to a broad audience of computational scientists, we provide instructive explanations of the underlying methodology, including overviews of the most commonly used deep learning architectures and feature representations of molecular data. We highlight the latest AI-based research that has led to the successful discovery of GPCR bioactive ligands. However, an equal focus of this review is on the discussion of machine learning-based technology that has been applied to ligand discovery in general and has the potential to pave the way for successful GPCR bioactive ligand discovery in the future. This review concludes with a brief outlook highlighting the recent research trends in deep learning, such as active learning and semi-supervised learning, which have great potential for advancing bioactive ligand discovery.

QSAR-derived affinity fingerprints (part 1): fingerprint construction and modeling performance for similarity searching, bioactivity classification and scaffold hopping

An affinity fingerprint is the vector consisting of compound’s affinity or potency against the reference panel of protein targets. Here, we present the QAFFP fingerprint, 440 elements long in silico QSAR-based affinity fingerprint, components of which are predicted by Random Forest regression models trained on bioactivity data from the ChEMBL database. Both real-valued (rv-QAFFP) and binary (b-QAFFP) versions of the QAFFP fingerprint were implemented and their performance in similarity searching, biological activity classification and scaffold hopping was assessed and compared to that of the 1024 bits long Morgan2 fingerprint (the RDKit implementation of the ECFP4 fingerprint). In both similarity searching and biological activity classification, the QAFFP fingerprint yields retrieval rates, measured by AUC (~ 0.65 and ~ 0.70 for similarity searching depending on data sets, and ~ 0.85 for classification) and EF5 (~ 4.67 and ~ 5.82 for similarity searching depending on data sets, and ~ 2.10 for classification), comparable to that of the Morgan2 fingerprint (similarity searching AUC of ~ 0.57 and ~ 0.66, and EF5 of ~ 4.09 and ~ 6.41, depending on data sets, classification AUC of ~ 0.87, and EF5 of ~ 2.16). However, the QAFFP fingerprint outperforms the Morgan2 fingerprint in scaffold hopping as it is able to retrieve 1146 out of existing 1749 scaffolds, while the Morgan2 fingerprint reveals only 864 scaffolds.

Using Predicted Bioactivity Profiles to Improve Predictive Modeling

Predictive modeling is a cornerstone in early drug development. Using information for multiple domains or across prediction tasks has the potential to improve the performance of predictive modeling. However, aggregating data often leads to incomplete data matrices that might be limiting for modeling. In line with previous studies, we show that by generating predicted bioactivity profiles, and using these as additional features, prediction accuracy of biological endpoints can be improved. Using conformal prediction, a type of confidence predictor, we present a robust framework for the calculation of these profiles and the evaluation of their impact. We report on the outcomes from several approaches to generate the predicted profiles on 16 datasets in cytotoxicity and bioactivity and show that efficiency is improved the most when including the p-values from conformal prediction as bioactivity profiles.

Machine learning applications in drug development

Due to the huge amount of biological and medical data available today, along with well-established machine learning algorithms, the design of largely automated drug development pipelines can now be envisioned. These pipelines may guide, or speed up, drug discovery; provide a better understanding of diseases and associated biological phenomena; help planning preclinical wet-lab experiments, and even future clinical trials. This automation of the drug development process might be key to the current issue of low productivity rate that pharmaceutical companies currently face. In this survey, we will particularly focus on two classes of methods: sequential learning and recommender systems, which are active biomedical fields of research.

Applications of machine learning in drug discovery and development

Drug discovery and development pipelines are long, complex and depend on numerous factors. Machine learning (ML) approaches provide a set of tools that can improve discovery and decision making for well-specified questions with abundant, high-quality data. Opportunities to apply ML occur in all stages of drug discovery. Examples include target validation, identification of prognostic biomarkers and analysis of digital pathology data in clinical trials. Applications have ranged in context and methodology, with some approaches yielding accurate predictions and insights. The challenges of applying ML lie primarily with the lack of interpretability and repeatability of ML-generated results, which may limit their application. In all areas, systematic and comprehensive high-dimensional data still need to be generated. With ongoing efforts to tackle these issues, as well as increasing awareness of the factors needed to validate ML approaches, the application of ML can promote data-driven decision making and has the potential to speed up the process and reduce failure rates in drug discovery and development.

Applications of Deep-Learning in Exploiting Large-Scale and Heterogeneous Compound Data in Industrial Pharmaceutical Research

In recent years, the development of high-throughput screening (HTS) technologies and their establishment in an industrialized environment have given scientists the possibility to test millions of molecules and profile them against a multitude of biological targets in a short period of time, generating data in a much faster pace and with a higher quality than before. Besides the structure activity data from traditional bioassays, more complex assays such as transcriptomics profiling or imaging have also been established as routine profiling experiments thanks to the advancement of Next Generation Sequencing or automated microscopy technologies. In industrial pharmaceutical research, these technologies are typically established in conjunction with automated platforms in order to enable efficient handling of screening collections of thousands to millions of compounds. To exploit the ever-growing amount of data that are generated by these approaches, computational techniques are constantly evolving. In this regard, artificial intelligence technologies such as deep learning and machine learning methods play a key role in cheminformatics and bio-image analytics fields to address activity prediction, scaffold hopping, de novo molecule design, reaction/retrosynthesis predictions, or high content screening analysis. Herein we summarize the current state of analyzing large-scale compound data in industrial pharmaceutical research and describe the impact it has had on the drug discovery process over the last two decades, with a specific focus on deep-learning technologies.

Combining structural and bioactivity-based fingerprints improves prediction performance and scaffold hopping capability

This study aims at improving upon existing activity predictions methods by augmenting chemical structure fingerprints with bio-activity based fingerprints derived from high-throughput screening (HTS) data (HTSFPs) and thereby showcasing the benefits of combining different descriptor types. This type of descriptor would be applied in an iterative screening scenario for more targeted compound set selection. The HTSFPs were generated from HTS data obtained from PubChem and combined with an ECFP4 structural fingerprint. The bioactivity-structure hybrid (BaSH) fingerprint was benchmarked against the individual ECFP4 and HTSFP fingerprints. Their performance was evaluated via retrospective analysis of a subset of the PubChem HTS data. Results showed that the BaSH fingerprint has improved predictive performance as well as scaffold hopping capability. The BaSH fingerprint identified unique compounds compared to both the ECFP4 and the HTSFP fingerprint indicating synergistic effects between the two fingerprints. A feature importance analysis showed that a small subset of the HTSFP features contribute most to the overall performance of the BaSH fingerprint. This hybrid approach allows for activity prediction of compounds with only sparse HTSFPs due to the supporting effect from the structural fingerprint.

Predicting kinase inhibitors using bioactivity matrix derived informer sets

Prediction of compounds that are active against a desired biological target is a common step in drug discovery efforts. Virtual screening methods seek some active-enriched fraction of a library for experimental testing. Where data are too scarce to train supervised learning models for compound prioritization, initial screening must provide the necessary data. Commonly, such an initial library is selected on the basis of chemical diversity by some pseudo-random process (for example, the first few plates of a larger library) or by selecting an entire smaller library. These approaches may not produce a sufficient number or diversity of actives. An alternative approach is to select an informer set of screening compounds on the basis of chemogenomic information from previous testing of compounds against a large number of targets. We compare different ways of using chemogenomic data to choose a small informer set of compounds based on previously measured bioactivity data. We develop this Informer-Based-Ranking (IBR) approach using the Published Kinase Inhibitor Sets (PKIS) as the chemogenomic data to select the informer sets. We test the informer compounds on a target that is not part of the chemogenomic data, then predict the activity of the remaining compounds based on the experimental informer data and the chemogenomic data. Through new chemical screening experiments, we demonstrate the utility of IBR strategies in a prospective test on three kinase targets not included in the PKIS.

In the early stages of drug discovery efforts, computational models are used to predict activity and prioritize compounds for experimental testing. New targets commonly lack the data necessary to build effective models, and the screening needed to generate that experimental data can be costly. We seek to improve the efficiency of the initial screening phase, and of the process of prioritizing compounds for subsequent screening.

We choose a small informer set of compounds based on publicly available prior screening data on distinct targets. We then collect experimental data on these informer compounds and use that data to predict the activity of other compounds in the set for the target of interest. Computational and statistical tools are needed to identify informer compounds and to prioritize other compounds for subsequent phases of screening. We find that selection of informer compounds on the basis of bioactivity data from previous screening efforts is superior to the traditional approach of selection of a chemically diverse subset of compounds. We demonstrate the success of this approach in retrospective tests on the Published Kinase Inhibitor Sets (PKIS) chemogenomic data and in prospective experimental screens against three additional non-human kinase targets.

Using Machine Learning to Predict Synergistic Antimalarial Compound Combinations With Novel Structures

The parasite Plasmodium falciparum is the most lethal species of Plasmodium to cause serious malaria infection in humans, and with resistance developing rapidly novel treatment modalities are currently being sought, one of which being combinations of existing compounds. The discovery of combinations of antimalarial drugs that act synergistically with one another is hence of great importance; however an exhaustive experimental screen of large drug space in a pairwise manner is not an option. In this study we apply our machine learning approach, Combination Synergy Estimation (CoSynE), which can predict novel synergistic drug interactions using only prior experimental combination screening data and knowledge of compound molecular structures, to a dataset of 1,540 antimalarial drug combinations in which 22.2% were synergistic. Cross validation of our model showed that synergistic CoSynE predictions are enriched 2.74 × compared to random selection when both compounds in a predicted combination are known from other combinations among the training data, 2.36 × when only one compound is known from the training data, and 1.5 × for entirely novel combinations. We prospectively validated our model by making predictions for 185 combinations of 23 entirely novel compounds. CoSynE predicted 20 combinations to be synergistic, which was experimentally validated for nine of them (45%), corresponding to an enrichment of 1.70 × compared to random selection from this prospective data set. Such enrichment corresponds to a 41% reduction in experimental effort. Interestingly, we found that pairwise screening of the compounds CoSynE individually predicted to be synergistic would result in an enrichment of 1.36 × compared to random selection, indicating that synergy among compound combinations is not a random event. The nine novel and correctly predicted synergistic compound combinations mainly (where sufficient bioactivity information is available) consist of efflux or transporter inhibitors (such as hydroxyzine), combined with compounds exhibiting antimalarial activity alone (such as sorafenib, apicidin, or dihydroergotamine). However, not all compound synergies could be rationalized easily in this way. Overall, this study highlights the potential for predictive modeling to expedite the discovery of novel drug combinations in fight against antimalarial resistance, while the underlying approach is also generally applicable.

Data-driven approaches used for compound library design, hit triage and bioactivity modeling in high-throughput screening

High-throughput screening (HTS) campaigns are routinely performed in pharmaceutical companies to explore activity profiles of chemical libraries for the identification of promising candidates for further investigation. With the aim of improving hit rates in these campaigns, data-driven approaches have been used to design relevant compound screening collections, enable effective hit triage and perform activity modeling for compound prioritization. Remarkable progress has been made in the activity modeling area since the recent introduction of large-scale bioactivity-based compound similarity metrics. This is evidenced by increased hit rates in iterative screening strategies and novel insights into compound mode of action obtained through activity modeling. Here, we provide an overview of the developments in data-driven approaches, elaborate on novel activity modeling techniques and screening paradigms explored and outline their significance in HTS.

Prospective Assessment of Virtual Screening Heuristics Derived Using a Novel Fusion Score

High-throughput screening (HTS) is a widespread method in early drug discovery for identifying promising chemical matter that modulates a target or phenotype of interest. Because HTS campaigns involve screening millions of compounds, it is often desirable to initiate screening with a subset of the full collection. Subsequently, virtual screening methods prioritize likely active compounds in the remaining collection in an iterative process. With this approach, orthogonal virtual screening methods are often applied, necessitating the prioritization of hits from different approaches. Here, we introduce a novel method of fusing these prioritizations and benchmark it prospectively on 17 screening campaigns using virtual screening methods in three descriptor spaces. We found that the fusion approach retrieves 15% to 65% more active chemical series than any single machine-learning method and that appropriately weighting contributions of similarity and machine-learning scoring techniques can increase enrichment by 1% to 19%. We also use fusion scoring to evaluate the tradeoff between screening more chemical matter initially in lieu of replicate samples to prevent false-positives and find that the former option leads to the retrieval of more active chemical series. These results represent guidelines that can increase the rate of identification of promising active compounds in future iterative screens.

Iterative Focused Screening with Biological Fingerprints Identifies Selective Asc-1 Inhibitors Distinct from Traditional High Throughput Screening

N-methyl-d-aspartate receptors (NMDARs) mediate glutamatergic signaling that is critical to cognitive processes in the central nervous system, and NMDAR hypofunction is thought to contribute to cognitive impairment observed in both schizophrenia and Alzheimer's disease. One approach to enhance the function of NMDAR is to increase the concentration of an NMDAR coagonist, such as glycine or d-serine, in the synaptic cleft. Inhibition of alanine-serine-cysteine transporter-1 (Asc-1), the primary transporter of d-serine, is attractive because the transporter is localized to neurons in brain regions critical to cognitive function, including the hippocampus and cortical layers III and IV, and is colocalized with d-serine and NMDARs. To identify novel Asc-1 inhibitors, two different screening approaches were performed with whole-cell amino acid uptake in heterologous cells stably expressing human Asc-1: (1) a high-throughput screen (HTS) of 3 M compounds measuring ³⁵S l-cysteine uptake into cells attached to scintillation proximity assay beads in a 1536 well format and (2) an iterative focused screen (IFS) of a 45 000 compound diversity set using a ³H d-serine uptake assay with a liquid scintillation plate reader in a 384 well format. Critically important for both screening approaches was the implementation of counter screens to remove nonspecific inhibitors of radioactive amino acid uptake. Furthermore, a 15 000 compound expansion step incorporating both on- and off-target data into chemical and biological fingerprint-based models for selection of additional hits enabled the identification of novel Asc-1-selective chemical matter from the IFS that was not identified in the full-collection HTS.

Profiling Prediction of Kinase Inhibitors: Toward the Virtual Assay

Kinome-wide screening would have the advantage of providing structure-activity relationships against hundreds of targets simultaneously. Here, we report the generation of ligand-based activity prediction models for over 280 kinases by employing Machine Learning methods on an extensive data set of proprietary bioactivity data combined with open data. High quality (AUC > 0.7) was achieved for ∼200 kinases by (1) combining open with proprietary data, (2) choosing Random Forest over alternative tested Machine Learning methods, and (3) balancing the training data sets. Tests on left-out and external data indicate a high value for virtual screening projects. Importantly, the derived models are evenly distributed across the kinome tree, allowing reliable profiling prediction for all kinase branches. The prediction quality was further improved by employing experimental bioactivity fingerprints of a small kinase subset. Overall, the generated models can support various hit identification tasks, including virtual screening, compound repurposing, and the detection of potential off-targets.

Aggregated Compound Biological Signatures Facilitate Phenotypic Drug Discovery and Target Elucidation

Predicting the cellular response of compounds is a challenge central to the discovery of new drugs. Compound biological signatures have risen as a way of representing the perturbation produced by a compound in the cell. However, their ability to encode specific phenotypic information and generating tangible predictions remains unknown, mainly because of the inherent noise in such data sets. In this work, we statistically aggregate signals from several compound biological signatures to find compounds that produce a desired phenotype in the cell. We exploit this method in two applications relevant for phenotypic screening in drug discovery programs: target-independent hit expansion and target identification. As a result, we present here (i) novel nanomolar inhibitors of cellular division that reproduce the phenotype and the mode of action of reference natural products and (ii) blockers of the NKCC1 cotransporter for autism spectrum disorders. Our results were confirmed in both cellular and biochemical assays of the respective projects. In addition, these examples provided novel insights on the information content and biological significance of compound biological signatures from HTS, and their applicability to drug discovery in general. For target identification, we show that novel targets can be predicted successfully for drugs by reporting new activities for nimedipine, fluspirilene, and pimozide and providing a rationale for repurposing and side effects. Our results highlight the opportunities of reusing public bioactivity data for prospective drug discovery, including scenarios where the effective target or mode of action of a particular molecule is not known, such as in phenotypic screening campaigns.

Filtered circular fingerprints improve either prediction or runtime performance while retaining interpretability

BACKGROUND

Even though circular fingerprints have been first introduced more than 50 years ago, they are still widely used for building highly predictive, state-of-the-art (Q)SAR models. Historically, these structural fragments were designed to search large molecular databases. Hence, to derive a compact representation, circular fingerprint fragments are often folded to comparatively short bit-strings. However, folding fingerprints introduces bit collisions, and therefore adds noise to the encoded structural information and removes its interpretability. Both representations, folded as well as unprocessed fingerprints, are often used for (Q)SAR modeling.

RESULTS

We show that it can be preferable to build (Q)SAR models with circular fingerprint fragments that have been filtered by supervised feature selection, instead of applying folded or all fragments. Compared to folded fingerprints, filtered fingerprints significantly increase predictive performance and remain unambiguous and interpretable. Compared to unprocessed fingerprints, filtered fingerprints reduce the computational effort and are a more compact and less redundant feature representation. Depending on the selected learning algorithm filtering yields about equally predictive (Q)SAR models. We demonstrate the suitability of filtered fingerprints for (Q)SAR modeling by presenting our freely available web service Collision-free Filtered Circular Fingerprints that provides rationales for predictions by highlighting important structural features in the query compound (see http://coffer.informatik.uni-mainz.de).

CONCLUSIONS

Circular fingerprints are potent structural features that yield highly predictive models and encode interpretable structural information. However, to not lose interpretability, circular fingerprints should not be folded when building prediction models. Our experiments show that filtering is a suitable option to reduce the high computational effort when working with all fingerprint fragments. Additionally, our experiments suggest that the area under precision recall curve is a more sensible statistic for validating (Q)SAR models for virtual screening than the area under ROC or other measures for early recognition.

GRAPHICAL ABSTRACT

MetMaxStruct: A Tversky-Similarity-Based Strategy for Analysing the (Sub)Structural Similarities of Drugs and Endogenous Metabolites

BACKGROUND

Previous studies compared the molecular similarity of marketed drugs and endogenous human metabolites (endogenites), using a series of fingerprint-type encodings, variously ranked and clustered using the Tanimoto (Jaccard) similarity coefficient (TS). Because this gives equal weight to all parts of the encoding (thence to different substructures in the molecule) it may not be optimal, since in many cases not all parts of the molecule will bind to their macromolecular targets. Unsupervised methods cannot alone uncover this. We here explore the kinds of differences that may be observed when the TS is replaced-in a manner more equivalent to semi-supervised learning-by variants of the asymmetric Tversky (TV) similarity, that includes α and β parameters.

RESULTS

Dramatic differences are observed in (i) the drug-endogenite similarity heatmaps, (ii) the cumulative "greatest similarity" curves, and (iii) the fraction of drugs with a Tversky similarity to a metabolite exceeding a given value when the Tversky α and β parameters are varied from their Tanimoto values. The same is true when the sum of the α and β parameters is varied. A clear trend toward increased endogenite-likeness of marketed drugs is observed when α or β adopt values nearer the extremes of their range, and when their sum is smaller. The kinds of molecules exhibiting the greatest similarity to two interrogating drug molecules (chlorpromazine and clozapine) also vary in both nature and the values of their similarity as α and β are varied. The same is true for the converse, when drugs are interrogated with an endogenite. The fraction of drugs with a Tversky similarity to a molecule in a library exceeding a given value depends on the contents of that library, and α and β may be "tuned" accordingly, in a semi-supervised manner. At some values of α and β drug discovery library candidates or natural products can "look" much more like (i.e., have a numerical similarity much closer to) drugs than do even endogenites.

CONCLUSIONS

Overall, the Tversky similarity metrics provide a more useful range of examples of molecular similarity than does the simpler Tanimoto similarity, and help to draw attention to molecular similarities that would not be recognized if Tanimoto alone were used. Hence, the Tversky similarity metrics are likely to be of significant value in many general problems in cheminformatics.

Data-Driven Derivation of an "Informer Compound Set" for Improved Selection of Active Compounds in High-Throughput Screening

Despite the usefulness of high-throughput screening (HTS) in drug discovery, for some systems, low assay throughput or high screening cost can prohibit the screening of large numbers of compounds. In such cases, iterative cycles of screening involving active learning (AL) are employed, creating the need for smaller "informer sets" that can be routinely screened to build predictive models for selecting compounds from the screening collection for follow-up screens. Here, we present a data-driven derivation of an informer compound set with improved predictivity of active compounds in HTS, and we validate its benefit over randomly selected training sets on 46 PubChem assays comprising at least 300,000 compounds and covering a wide range of assay biology. The informer compound set showed improvement in BEDROC(α = 100), PRAUC, and ROCAUC values averaged over all assays of 0.024, 0.014, and 0.016, respectively, compared to randomly selected training sets, all with paired t-test p-values <10(-15). A per-assay assessment showed that the BEDROC(α = 100), which is of particular relevance for early retrieval of actives, improved for 38 out of 46 assays, increasing the success rate of smaller follow-up screens. Overall, we showed that an informer set derived from historical HTS activity data can be employed for routine small-scale exploratory screening in an assay-agnostic fashion. This approach led to a consistent improvement in hit rates in follow-up screens without compromising scaffold retrieval. The informer set is adjustable in size depending on the number of compounds one intends to screen, as performance gains are realized for sets with more than 3,000 compounds, and this set is therefore applicable to a variety of situations. Finally, our results indicate that random sampling may not adequately cover descriptor space, drawing attention to the importance of the composition of the training set for predicting actives.

Assessment of the classification abilities of the CNS multi-parametric optimization approach by the method of logistic regression

Assessment of "CNS drugs/CNS candidates" classification abilities of the multi-parametric optimization (CNS MPO) approach was performed by logistic regression. It was found that the five out of the six separately used physical-chemical properties (topological polar surface area, number of hydrogen-bonded donor atoms, basicity, lipophilicity of compound in neutral form and at pH = 7.4) provided accuracy of recognition below 60%. Only the descriptor of molecular weight (MW) could correctly classify two-thirds of the studied compounds. Aggregation of all six properties in the MPOscore did not improve the classification, which was worse than the classification using only MW. The results of our study demonstrate the imperfection of the CNS MPO approach; in its current form it is not very useful for computer design of new, effective CNS drugs.

Evidence-Based and Quantitative Prioritization of Tool Compounds in Phenotypic Drug Discovery

The use of potent and selective chemical tools with well-defined targets can help elucidate biological processes driving phenotypes in phenotypic screens. However, identification of selective compounds en masse to create targeted screening sets is non-trivial. A systematic approach is needed to prioritize probes, which prevents the repeated use of published but unselective compounds. Here we performed a meta-analysis of integrated large-scale, heterogeneous bioactivity data to create an evidence-based, quantitative metric to systematically rank tool compounds for targets. Our tool score (TS) was then tested on hundreds of compounds by assessing their activity profiles in a panel of 41 cell-based pathway assays. We demonstrate that high-TS tools show more reliably selective phenotypic profiles than lower-TS compounds. Additionally we highlight frequently tested compounds that are non-selective tools and distinguish target family polypharmacology from cross-family promiscuity. TS can therefore be used to prioritize compounds from heterogeneous databases for phenotypic screening.

bioassayR: Cross-Target Analysis of Small Molecule Bioactivity

Despite a large and rapidly growing body of small molecule bioactivity screens available in the public domain, systematic leverage of the data to assess target druggability and compound selectivity has been confounded by a lack of suitable cross-target analysis software. We have developed bioassayR, a computational tool that enables simultaneous analysis of thousands of bioassay experiments performed over a diverse set of compounds and biological targets. Unique features include support for large-scale cross-target analyses of both public and custom bioassays, generation of high throughput screening fingerprints (HTSFPs), and an optional preloaded database that provides access to a substantial portion of publicly available bioactivity data. bioassayR is implemented as an open-source R/Bioconductor package available from https://bioconductor.org/packages/bioassayR/.

Analysis of Iterative Screening with Stepwise Compound Selection Based on Novartis In-house HTS Data

With increased automation and larger compound collections, the development of high-throughput screening (HTS) started replacing previous approaches in drug discovery from around the 1980s onward. However, even today it is not always appropriate, or even feasible, to screen large collections of compounds in a particular assay. Here, we present an efficient method for iterative screening of small subsets of compound libraries. With this method, the retrieval of active compounds is optimized using their structural information and biological activity fingerprints. We validated this approach retrospectively on 34 Novartis in-house HTS assays covering a wide range of assay biology, including cell proliferation, antibacterial activity, gene expression, and phosphorylation. This method was employed to retrieve subsets of compounds for screening, where selected hits from any given round of screening were used as starting points to select chemically and biologically similar compounds for the next iteration. By only screening ∼1% of the full screening collection (∼15 000 compounds), the method consistently retrieves diverse compounds belonging to the top 0.5% of the most active compounds for the HTS campaign. For most of the assays, over half of the compounds selected by the method were found to be among the 5% most active compounds of the corresponding full-deck HTS. In addition, the stringency of the iterative method can be modified depending on the number of compounds one can afford to screen, making it a flexible tool to discover active compounds efficiently.