Search

Funder: FP7 People: Marie-Curie Actions; doi: http://dx.doi.org/10.13039/100011264; Grant(s): 238701, 238701, 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.

Animal pharmacokinetic (PK) data as well as human and animal in vitro systems are utilized in drug discovery to define the rate and route of drug elimination. Accurate prediction and mechanistic understanding of drug clearance and disposition in animals provide a degree of confidence for extrapolation to humans. In addition, prediction of in vivo properties can be used to improve design during drug discovery, help select compounds with better properties, and reduce the number of in vivo experiments. In this study, we generated machine learning models able to predict rat in vivo PK parameters and concentration-time PK profiles based on the molecular chemical structure and either measured or predicted in vitro parameters. The models were trained on internal in vivo rat PK data for over 3000 diverse compounds from multiple projects and therapeutic areas, and the predicted endpoints include clearance and oral bioavailability. We compared the performance of various traditional machine learning algorithms and deep learning approaches, including graph convolutional neural networks. The best models for PK parameters achieved R2 = 0.63 [root mean squared error (RMSE) = 0.26] for clearance and R2 = 0.55 (RMSE = 0.46) for bioavailability. The models provide a fast and cost-efficient way to guide the design of molecules with optimal PK profiles, to enable the prediction of virtual compounds at the point of design, and to drive prioritization of compounds for in vivo assays.

Gene expression and cell morphology data are high-dimensional biological readouts of much recent interest for drug discovery. They are able to describe biological systems in different states (e.g., healthy and diseased), as well as biological systems before and after compound treatment, and they are hence useful for matching both spaces (e.g., for drug repurposing) as well as for characterizing compounds with respect to efficacy and safety endpoints. This Microperspective describes recent advances in this direction with a focus on applied drug discovery and drug repurposing, as well as outlining what else is needed to advance further, with a particular focus on better understanding the applicability domain of readouts and their relevance for decision making, which is currently often still unclear.

Background A key histopathological hallmark of Alzheimer’s disease (AD) is the presence of neurofibrillary tangles of aggregated microtubule-associated protein tau in neurons. Anle138b is a small molecule which has previously shown efficacy in mice in reducing tau aggregates and rescuing AD disease phenotypes. Methods In this work, we employed bioinformatics analysis—including pathway enrichment and causal reasoning—of an in vitro tauopathy model. The model consisted of cultured rat cortical neurons either unseeded or seeded with tau aggregates derived from human AD patients, both of which were treated with Anle138b to generate hypotheses for its mode of action. In parallel, we used a collection of human target prediction models to predict direct targets of Anle138b based on its chemical structure. Results Combining the different approaches, we found evidence supporting the hypothesis that the action of Anle138b involves several processes which are key to AD progression, including cholesterol homeostasis and neuroinflammation. On the pathway level, we found significantly enriched pathways related to these two processes including those entitled “Superpathway of cholesterol biosynthesis” and “Granulocyte adhesion and diapedesis”. With causal reasoning, we inferred differential activity of SREBF1/2 (involved in cholesterol regulation) and mediators of the inflammatory response such as NFKB1 and RELA. Notably, our findings were also observed in Anle138b-treated unseeded neurons, meaning that the inferred processes are independent of tau pathology and thus represent the direct action of the compound in the cellular system. Through structure-based ligand-target prediction, we predicted the intracellular cholesterol carrier NPC1 as well as NF-κB subunits as potential targets of Anle138b, with structurally similar compounds in the model training set known to target the same proteins. Conclusions This study has generated feasible hypotheses for the potential mechanism of action of Anle138b, which will enable the development of future molecular interventions aiming to reduce tau pathology in AD patients.

Mitochondrial toxicity is an important safety endpoint in drug discovery. Models based solely on chemical structure for predicting mitochondrial toxicity are currently limited in accuracy and applicability domain to the chemical space of the training compounds. In this work, we aimed to utilize both -omics and chemical data to push beyond the state-of-the-art. We combined Cell Painting and Gene Expression data with chemical structural information from Morgan fingerprints for 382 chemical perturbants tested in the Tox21 mitochondrial membrane depolarization assay. We observed that mitochondrial toxicants differ from non-toxic compounds in morphological space and identified compound clusters having similar mechanisms of mitochondrial toxicity, thereby indicating that morphological space provides biological insights related to mechanisms of action of this endpoint. We further showed that models combining Cell Painting, Gene Expression features and Morgan fingerprints improved model performance on an external test set of 244 compounds by 60% (in terms of F1 score) and improved extrapolation to new chemical space. The performance of our combined models was comparable with dedicated in vitro assays for mitochondrial toxicity. Our results suggest that combining chemical descriptors with biological readouts enhances the detection of mitochondrial toxicants, with practical implications in drug discovery.Abstract FigureGraphical Abstract

Funder: Science and Technology Facilities Council, Funder: Wellcome Trust, The field of machine learning for drug discovery is witnessing an explosion of novel methods. These methods are often benchmarked on simple physicochemical properties such as solubility or general druglikeness, which can be readily computed. However, these properties are poor representatives of objective functions in drug design, mainly because they do not depend on the candidate compound's interaction with the target. By contrast, molecular docking is a widely applied method in drug discovery to estimate binding affinities. However, docking studies require a significant amount of domain knowledge to set up correctly, which hampers adoption. Here, we present dockstring, a bundle for meaningful and robust comparison of ML models using docking scores. dockstring consists of three components: (1) an open-source Python package for straightforward computation of docking scores, (2) an extensive dataset of docking scores and poses of more than 260,000 molecules for 58 medically relevant targets, and (3) a set of pharmaceutically relevant benchmark tasks such as virtual screening or de novo design of selective kinase inhibitors. The Python package implements a robust ligand and target preparation protocol that allows nonexperts to obtain meaningful docking scores. Our dataset is the first to include docking poses, as well as the first of its size that is a full matrix, thus facilitating experiments in multiobjective optimization and transfer learning. Overall, our results indicate that docking scores are a more realistic evaluation objective than simple physicochemical properties, yielding benchmark tasks that are more challenging and more closely related to real problems in drug discovery.

Funder: AstraZeneca, PROteolysis TArgeting Chimeras (PROTACs) use the ubiquitin-proteasome system to degrade a protein of interest for therapeutic benefit. Advances made in targeted protein degradation technology have been remarkable, with several molecules having moved into clinical studies. However, robust routes to assess and better understand the safety risks of PROTACs need to be identified, which is an essential step toward delivering efficacious and safe compounds to patients. In this work, we used Cell Painting, an unbiased high-content imaging method, to identify phenotypic signatures of PROTACs. Chemical clustering and model prediction allowed the identification of a mitotoxicity signature that could not be expected by screening the individual PROTAC components. The data highlighted the benefit of unbiased phenotypic methods for identifying toxic signatures and the potential to impact drug design.

Funder: Unilever, Estimation of points of departure (PoDs) from high-throughput transcriptomic data (HTTr) represents a key step in the development of next-generation risk assessment (NGRA). Current approaches mainly rely on single key gene targets, which are constrained by the information currently available in the knowledge base and make interpretation challenging as scientists need to interpret PoDs for thousands of genes or hundreds of pathways. In this work, we aimed to address these issues by developing a computational workflow to investigate the pathway concentration-response relationships in a way that is not fully constrained by known biology and also facilitates interpretation. We employed the Pathway-Level Information ExtractoR (PLIER) to identify latent variables (LVs) describing biological activity and then investigated in vitro LVs' concentration-response relationships using the ToxCast pipeline. We applied this methodology to a published transcriptomic concentration-response data set for 44 chemicals in MCF-7 cells and showed that our workflow can capture known biological activity and discriminate between estrogenic and antiestrogenic compounds as well as activity not aligning with the existing knowledge base, which may be relevant in a risk assessment scenario. Moreover, we were able to identify the known estrogen activity in compounds that are not well-established ER agonists/antagonists supporting the use of the workflow in read-across. Next, we transferred its application to chemical compounds tested in HepG2, HepaRG, and MCF-7 cells and showed that PoD estimates are in strong agreement with those estimated using a recently developed Bayesian approach (cor = 0.89) and in weak agreement with those estimated using a well-established approach such as BMDExpress2 (cor = 0.57). These results demonstrate the effectiveness of using PLIER in a concentration-response scenario to investigate pathway activity in a way that is not fully constrained by the knowledge base and to ease the biological interpretation and support the development of an NGRA framework with the ability to improve current risk assessment strategies for chemicals using new approach methodologies.

Summary Resistance to current therapies is common for pancreatic cancer and hence novel treatment options are urgently needed. In this work, we developed and validated a computational method to select synergistic compound combinations based on transcriptomic profiles from both the disease and compound side, combined with a pathway scoring system, which was then validated prospectively by testing 30 compounds (and their combinations) on PANC-1 cells. Some compounds selected as single agents showed lower GI50 values than the standard of care, gemcitabine. Compounds suggested as combination agents with standard therapy gemcitabine based on the best performing scoring system showed on average 2.82–5.18 times higher synergies compared to compounds that were predicted to be active as single agents. Examples of highly synergistic in vitro validated compound pairs include gemcitabine combined with Entinostat, thioridazine, loperamide, scriptaid and Saracatinib. Hence, the computational approach presented here was able to identify synergistic compound combinations against pancreatic cancer cells., Graphical abstract, Highlights • We developed a computational approach to predict synergistic compound pairs • The approach uses transcriptional data and pathway information for scoring • The predicted combinations with gemcitabine were validated in vitro • Predicted combinations versus single agents had 2.82–5.18 times higher synergies, Molecular biology; Computational bioinformatics; Cancer systems biology

Funder: H2020 European Research Council; doi: http://dx.doi.org/10.13039/100010663, Differentiation therapy is attracting increasing interest in cancer as it can be more specific than conventional chemotherapy approaches, and it has offered new treatment options for some cancer types, such as treating acute promyelocytic leukaemia (APL) by retinoic acid. However, there is a pressing need to identify additional molecules which act in this way, both in leukaemia and other cancer types. In this work, we hence developed a novel transcriptional drug repositioning approach, based on both bioinformatics and cheminformatics components, that enables selecting such compounds in a more informed manner. We have validated the approach for leukaemia cells, and retrospectively retinoic acid was successfully identified using our method. Prospectively, the anti-parasitic compound fenbendazole was tested in leukaemia cells, and we were able to show that it can induce the differentiation of leukaemia cells to granulocytes in low concentrations of 0.1 μM and within as short a time period as 3 days. This work hence provides a systematic and validated approach for identifying small molecules for differentiation therapy in cancer.

Affinity fingerprints report the activity of small molecules across a set of assays, and thus permit to gather information about the bioactivities of structurally dissimilar compounds, where models based on chemical structure alone are often limited, and model complex biological endpoints, such as human toxicity and in vitro cancer cell line sensitivity. Here, we propose to model in vitro compound activity using computationally predicted bioactivity profiles as compound descriptors. To this aim, we apply and validate a framework for the calculation of QSAR-derived affinity fingerprints (QAFFP) using a set of 1360 QSAR models generated using Ki, Kd, IC50 and EC50 data from ChEMBL database. QAFFP thus represent a method to encode and relate compounds on the basis of their similarity in bioactivity space. To benchmark the predictive power of QAFFP we assembled IC50 data from ChEMBL database for 18 diverse cancer cell lines widely used in preclinical drug discovery, and 25 diverse protein target data sets. This study complements part 1 where the performance of QAFFP in similarity searching, scaffold hopping, and bioactivity classification is evaluated. Despite being inherently noisy, we show that using QAFFP as descriptors leads to errors in prediction on the test set in the ~ 0.65–0.95 pIC50 units range, which are comparable to the estimated uncertainty of bioactivity data in ChEMBL (0.76–1.00 pIC50 units). We find that the predictive power of QAFFP is slightly worse than that of Morgan2 fingerprints and 1D and 2D physicochemical descriptors, with an effect size in the 0.02–0.08 pIC50 units range. Including QSAR models with low predictive power in the generation of QAFFP does not lead to improved predictive power. Given that the QSAR models we used to compute the QAFFP were selected on the basis of data availability alone, we anticipate better modeling results for QAFFP generated using more diverse and biologically meaningful targets. Data sets and Python code are publicly available at https://github.com/isidroc/QAFFP_regression.

Shexiang Baoxin Pill (SBP) is an oral formulation of Chinese materia medica for the treatment of angina pectoris. It displays pleiotropic roles in protecting the cardiovascular system. However, the mode of action of SBP in promoting angiogenesis, and in particular the synergy between its constituents is currently not fully understood. The combination of ginsenosides Rb2 and Rg3 were studied in human umbilical vein endothelial cells (HUVECs) for their proangiogenic effects. To understand the mode of action of the combination in more mechanistic detail, RNA-Seq analysis was conducted, and differentially expressed genes (DEGs), pathway analysis and Weighted Gene Correlation Network Analysis (WGCNA) were applied to further identify important genes that a play pivotal role in the combination treatment. The effects of pathway-specific inhibitors were observed to provide further support for the hypothesized mode of action of the combination. Ginsenosides Rb2 and Rg3 synergistically promoted HUVEC proliferation and tube formation under defined culture conditions. Also, the combination of Rb2/Rg3 rescued cells from homocysteine-induced damage. mRNA expression of CXCL8, CYR61, FGF16 and FGFRL1 was significantly elevated by the Rb2/Rg3 treatment, and representative signaling pathways induced by these genes were found. The increase of protein levels of phosphorylated-Akt and ERK42/44 by the Rb2/Rg3 combination supports the notion that it promotes endothelial cell proliferation via the PI3K/Akt and MAPK/ERK signaling pathways. The present study provides the hypothesis that SBP, via ginsenosides Rb2 and Rg3, involves the CXCR1/2 CXCL8 (IL8)-mediated PI3K/Akt and MAPK/ERK signaling pathways in achieving its proangiogenic effects.

To improve our ability to extrapolate preclinical toxicity to humans, there is a need to understand and quantify the concordance of adverse events (AEs) between animal models and clinical studies. In the present work, we discovered 3011 statistically significant associations between preclinical and clinical AEs caused by drugs reported in the PharmaPendium database of which 2952 were new associations between toxicities encoded by different Medical Dictionary for Regulatory Activities terms across species. To find plausible and testable candidate off-target drug activities for the derived associations, we investigated the genetic overlap between the genes linked to both a preclinical and a clinical AE and the protein targets found to interact with one or more drugs causing both AEs. We discuss three associations from the analysis in more detail for which novel candidate off-target drug activities could be identified, namely, the association of preclinical mutagenicity readouts with clinical teratospermia and ovarian failure, the association of preclinical reflexes abnormal with clinical poor-quality sleep, and the association of preclinical psychomotor hyperactivity with clinical drug withdrawal syndrome. Our analysis successfully identified a total of 77% of known safety targets currently tested in in vitro screening panels plus an additional 431 genes which were proposed for investigation as future safety targets for different clinical toxicities. This work provides new translational toxicity relationships beyond AE term-matching, the results of which can be used for risk profiling of future new chemical entities for clinical studies and for the development of future in vitro safety panels.

The application of convolutional neural networks (ConvNets) to harness high-content screening images or 2D compound representations is gaining increasing attention in drug discovery. However, existing applications often require large data sets for training, or sophisticated pretraining schemes. Here, we show using 33 IC50 data sets from ChEMBL 23 that the in vitro activity of compounds on cancer cell lines and protein targets can be accurately predicted on a continuous scale from their Kekulé structure representations alone by extending existing architectures (AlexNet, DenseNet-201, ResNet152 and VGG-19), which were pretrained on unrelated image data sets. We show that the predictive power of the generated models, which just require standard 2D compound representations as input, is comparable to that of Random Forest (RF) models and fully-connected Deep Neural Networks trained on circular (Morgan) fingerprints. Notably, including additional fully-connected layers further increases the predictive power of the ConvNets by up to 10%. Analysis of the predictions generated by RF models and ConvNets shows that by simply averaging the output of the RF models and ConvNets we obtain significantly lower errors in prediction for multiple data sets, although the effect size is small, than those obtained with either model alone, indicating that the features extracted by the convolutional layers of the ConvNets provide complementary predictive signal to Morgan fingerprints. Lastly, we show that multi-task ConvNets trained on compound images permit to model COX isoform selectivity on a continuous scale with errors in prediction comparable to the uncertainty of the data. Overall, in this work we present a set of ConvNet architectures for the prediction of compound activity from their Kekulé structure representations with state-of-the-art performance, that require no generation of compound descriptors or use of sophisticated image processing techniques. The code needed to reproduce the results presented in this study and all the data sets are provided at https://github.com/isidroc/kekulescope.

Adverse drug reactions (ADRs) are undesired effects of medicines that can harm patients and are a significant source of attrition in drug development. ADRs are anticipated by routinely screening drugs against secondary pharmacology protein panels. However, there is still a lack of quantitative information on the links between these off-target proteins and the risk of ADRs in humans. Here, we present a systematic analysis of associations between measured and predicted in vitro bioactivities of drugs, and adverse events (AEs) in humans from two sources of data: the Side Effect Resource (SIDER), derived from clinical trials, and the Food and Drug Administration Adverse Event Reporting System (FAERS), derived from post-marketing surveillance. The ratio of a drug’s in vitro potency against a given protein relative to its therapeutic unbound drug plasma concentration was used to select proteins most likely to be relevant to in vivo effects. In examining individual target bioactivities as predictors of AEs, we found a trade-off between the Positive Predictive Value and the fraction of drugs with AEs that can be detected, however considering sets of multiple targets for the same AE can help identify a greater fraction of AE-associated drugs. Of the 45 targets with statistically significant associations to AEs, 30 are included on existing safety target panels. The remaining 15 targets include 8 carbonic anhydrases, of which CA5B was significantly associated with cholestatic jaundice. We include the full quantitative data on associations between in vitro bioactivities and AEs in humans in this work, which can be used to make a more informed selection of safety profiling targets.

Motivation While drug combination therapies are a well-established concept in cancer treatment, identifying novel synergistic combinations is challenging due to the size of combinatorial space. However, computational approaches have emerged as a time- and cost-efficient way to prioritize combinations to test, based on recently available large-scale combination screening data. Recently, Deep Learning has had an impact in many research areas by achieving new state-of-the-art model performance. However, Deep Learning has not yet been applied to drug synergy prediction, which is the approach we present here, termed DeepSynergy. DeepSynergy uses chemical and genomic information as input information, a normalization strategy to account for input data heterogeneity, and conical layers to model drug synergies. Results DeepSynergy was compared to other machine learning methods such as Gradient Boosting Machines, Random Forests, Support Vector Machines and Elastic Nets on the largest publicly available synergy dataset with respect to mean squared error. DeepSynergy significantly outperformed the other methods with an improvement of 7.2% over the second best method at the prediction of novel drug combinations within the space of explored drugs and cell lines. At this task, the mean Pearson correlation coefficient between the measured and the predicted values of DeepSynergy was 0.73. Applying DeepSynergy for classification of these novel drug combinations resulted in a high predictive performance of an AUC of 0.90. Furthermore, we found that all compared methods exhibit low predictive performance when extrapolating to unexplored drugs or cell lines, which we suggest is due to limitations in the size and diversity of the dataset. We envision that DeepSynergy could be a valuable tool for selecting novel synergistic drug combinations. Availability and implementation DeepSynergy is available via www.bioinf.jku.at/software/DeepSynergy. Supplementary information Supplementary data are available at Bioinformatics online.

Functional magnetic resonance imaging (fMRI) is an extensively used method for the investigation of normal and pathological brain function. In particular, fMRI has been used to characterize spatiotemporal hemodynamic response to pharmacological challenges as a non-invasive readout of neuronal activity. However, the mechanisms underlying regional signal changes are yet unclear. In this study, we use a meta-analytic approach to converge data from microdialysis experiments with relative cerebral blood volume (rCBV) changes following acute administration of neuropsychiatric drugs in adult male rats. At whole-brain level, the functional response patterns show very weak correlation with neurochemical alterations, while for numerous brain areas a strong positive correlation with noradrenaline release exists. At a local scale of individual brain regions, the rCBV response to neurotransmitters is anatomically heterogeneous and, importantly, based on a complex interplay of different neurotransmitters that often exert opposing effects, thus providing a mechanism for regulating and fine tuning hemodynamic responses in specific regions.

Despite the increasing knowledge in both the chemical and biological domains the assimilation and exploration of heterogeneous datasets, encoding information about the chemical, bioactivity and phenotypic properties of compounds, remains a challenge due to requirement for overlap between chemicals assayed across the spaces. Here, we have constructed a novel dataset, larger than we have used in prior work, comprising 579 acute oral toxic compounds and 1427 non-toxic compounds derived from regulatory GHS information, along with their corresponding molecular and protein target descriptors and qHTS in vitro assay readouts from the Tox21 project. We found no clear association between the results of a FAFDrugs4 toxicophore screen and the acute oral toxicity classifications for our compound set; and a screen using a subset of the ToxAlerts toxicophores was also of limited utility, with only slight enrichment toward the toxic set (odds ratio of 1.48). We then investigated to what degree toxic and non-toxic compounds could be separated in each of the spaces, to compare their potential contribution to further analyses. Using an LDA projection, we found the largest degree of separation using chemical descriptors (Cohen’s d of 1.95) and the lowest degree of separation between toxicity classes using qHTS descriptors (Cohen’s d of 0.67). To compare the predictivity of the feature spaces for the toxicity endpoint, we next trained Random Forest (RF) acute oral toxicity classifiers on either molecular, protein target and qHTS descriptors. RFs trained on molecular and protein target descriptors were most predictive, with ROC AUC values of 0.80–0.92 and 0.70–0.85, respectively, across three test sets. RFs trained on both chemical and protein target descriptors combined exhibited similar predictive performance to the single-domain models (ROC AUC of 0.80–0.91). Model interpretability was improved by the inclusion of protein target descriptors, which allow the identification of specific targets (e.g. Retinal dehydrogenase) with literature links to toxic modes of action (e.g. oxidative stress). The dataset compiled in this study has been made available for future application. Electronic supplementary material The online version of this article (10.1186/s13321-019-0356-5) contains supplementary material, which is available to authorized users.

The rapid emergence of pesticide resistance has given rise to a demand for herbicides with new mode of action (MoA). In the agrochemical sector, with the availability of experimental high throughput screening (HTS) data, it is now possible to utilize in silico target prediction methods in the early discovery phase to suggest the MoA of a compound via data mining of bioactivity data. While having been established in the pharmaceutical context, in the agrochemical area this approach poses rather different challenges, as we have found in this work, partially due to different chemistry, but even more so due to different (usually smaller) amounts of data, and different ways of conducting HTS. With the aim to apply computational methods for facilitating herbicide target identification, 48,000 bioactivity data against 16 herbicide targets were processed to train Laplacian modified Naïve Bayesian (NB) classification models. The herbicide target prediction model ("HerbiMod") is an ensemble of 16 binary classification models which are evaluated by internal, external and prospective validation sets. In addition to the experimental inactives, 10,000 random agrochemical inactives were included in the training process, which showed to improve the overall balanced accuracy of our models up to 40%. For all the models, performance in terms of balanced accuracy of≥80% was achieved in five-fold cross validation. Ranking target predictions was addressed by means of z-scores which improved predictivity over using raw scores alone. An external testset of 247 compounds from ChEMBL and a prospective testset of 394 compounds from BASF SE tested against five well studied herbicide targets (ACC, ALS, HPPD, PDS and PROTOX) were used for further validation. Only 4% of the compounds in the external testset lied in the applicability domain and extrapolation (and correct prediction) was hence impossible, which on one hand was surprising, and on the other hand illustrated the utilization of using applicability domains in the first place. However, performance better than 60% in balanced accuracy was achieved on the prospective testset, where all the compounds fell within the applicability domain, and which hence underlines the possibility of using target prediction also in the area of agrochemicals.

The epidermal growth factor receptor (EGFR) is a validated therapeutic target for triple-negative breast cancer (TNBC). In the present study, we synthesize novel adamantanyl-based thiadiazolyl pyrazoles by introducing the adamantane ring to thiazolopyrazoline. On the basis of loss of cell viability in TNBC cells, 4-(adamantan-1-yl)-2-(3-(2,4-dichlorophenyl)-5-phenyl-4,5-dihydro-1H-pyrazol-1-yl)thiazole (APP) was identified as a lead compound. Using a Parzen-Rosenblatt Window classifier, APP was predicted to target the EGFR protein, and the same was confirmed by surface plasmon resonance. Further analysis revealed that APP suppressed the phosphorylation of EGFR at Y992, Y1045, Y1068, Y1086, Y1148, and Y1173 in TNBC cells. APP also inhibited the phosphorylation of ERK at Y204 and of STAT3 at Y705, implying that APP downregulates the activity of EGFR downstream effectors. Small interfering RNA mediated depletion of EGFR expression prevented the effect of APP in BT549 and MDA-MB-231 cells, indicating that APP specifically targets the EGFR. Furthermore, APP modulated the expression of the proteins involved in cell proliferation and survival. In addition, APP altered the expression of epithelial-mesenchymal transition related proteins and suppressed the invasion of TNBC cells. Hence, we report a novel and specific inhibitor of the EGFR signaling cascade.

The development of treatments involving combinations of drugs is a promising approach towards combating complex or multifactorial disorders. However, the large number of compound combinations that can be generated, even from small compound collections, means that exhaustive experimental testing is infeasible. The ability to predict the behaviour of compound combinations in biological systems, whittling down the number of combinations to be tested, is therefore crucial. Here, we review the current state-of-the-art in the field of compound combination modelling, with the aim to support the development of approaches that, as we hope, will finally lead to an integration of chemical with systems-level biological information for predicting the effect of chemical mixtures.

As the popularity of transcriptomic analysis has grown, the reported lack of concordance between different studies of the same condition has become a growing concern, raising questions as to the representativeness of different study types, such as non-human disease models or studies of surrogate tissues, to gene expression in the human condition.In a comparison of 33 microarray studies of Parkinson's disease, correlation and clustering analyses were used to determine the factors influencing concordance between studies, including agreement between different tissue types, different microarray platforms, and between neurotoxic and genetic disease models and human Parkinson's disease.Concordance over all studies is low, with correlation of only 0.05 between differential gene expression signatures on average, but increases within human patients and studies of the same tissue type, rising to 0.38 for studies of human substantia nigra. Agreement of animal models, however, is dependent on model type. Studies of brain tissue from Parkinson's disease patients (specifically the substantia nigra) form a distinct group, showing patterns of differential gene expression noticeably different from that in non-brain tissues and animal models of Parkinson's disease; while comparison with other brain diseases (Alzheimer's disease and brain cancer) suggests that the mixed study types display a general signal of neurodegenerative disease. A meta-analysis of these 33 microarray studies demonstrates the greater ability of studies in humans and highly-affected tissues to identify genes previously known to be associated with Parkinson's disease.The observed clustering and concordance results suggest the existence of a 'characteristic' signal of Parkinson's disease found in significantly affected human tissues in humans. These results help to account for the consistency (or lack thereof) so far observed in microarray studies of Parkinson's disease, and act as a guide to the selection of transcriptomic studies most representative of the underlying gene expression changes in the human disease.

Selective modulators of the γ-amino butyric acid (GABAA) family of receptors have the potential to treat a range of disease states related to cognition, pain, and anxiety. While the development of various α subunit-selective modulators is currently underway for the treatment of anxiety disorders, a mechanistic understanding of the correlation between their bioactivity and efficacy, based on ligand-target interactions, is currently still lacking. In order to alleviate this situation, in the current study we have analyzed, using ligand- and structure-based methods, a data set of 5440 GABAA modulators. The Spearman correlation (ρ) between binding activity and efficacy of compounds was calculated to be 0.008 and 0.31 against the α1 and α2 subunits of GABA receptor, respectively; in other words, the compounds had little diversity in structure and bioactivity, but they differed significantly in efficacy. Two compounds were selected as a case study for detailed interaction analysis due to the small difference in their structures and affinities (ΔpKi(comp1_α1 - comp2_α1) = 0.45 log units, ΔpKi(comp1_α2 - comp2_α2) = 0 log units) as compared to larger relative efficacies (ΔRE(comp1_α1 - comp2_α1) = 1.03, ΔRE(comp1_α2 - comp2_α2) = 0.21). Docking analysis suggested that His-101 is involved in a characteristic interaction of the α1 receptor with both compounds 1 and 2. Residues such as Phe-77, Thr-142, Asn-60, and Arg-144 of the γ chain of the α1γ2 complex also showed interactions with heterocyclic rings of both compounds 1 and 2, but these interactions were disturbed in the case of α2γ2 complex docking results. Binding pocket stability analysis based on molecular dynamics identified three substitutions in the loop C region of the α2 subunit, namely, G200E, I201T, and V202I, causing a reduction in the flexibility of α2 compared to α1. These amino acids in α2, as compared to α1, were also observed to decrease the vibrational and dihedral entropy and to increase the hydrogen bond content in α2 in the apo state. However, freezing of both α1 and α2 was observed in the ligand-bound state, with an increased number of internal hydrogen bonds and increased entropy. Therefore, we hypothesize that the amino acid differences in the loop C region of α2 are responsible for conformational changes in the protein structure compared to α1, as well as for the binding modes of compounds and hence their functional signaling.

The bromodomain-containing proteins are a ligandable family of epigenetic readers, which play important roles in oncological, cardiovascular, and inflammatory diseases. Achieving selective inhibition of specific bromodomains is challenging, due to the limited understanding of compound and target selectivity features. In this study we build and benchmark proteochemometric (PCM) classification models on bioactivity data for 15,350 data points across 31 bromodomains, using both compound fingerprints and binding site protein descriptors as input variables, achieving a maximum performance as measured by the Matthew's Correlation Coefficient (MCC) of 0.83 on the external test set. We also find that histone peptide binding data can be used as a target descriptor to build a high performing PCM model (MCC 0.80), showing the transferability of peptide interaction information to modeling small-molecule bioactivity. 1,139 compounds were selected for prospective experimental testing by performing a virtual screen using model predictions and implementing conformal prediction, which resulted in 319 correctly predicted compound-target pair actives and the correct prediction for certain selectivity profile combinations of the four bromodomains tested against. We identify that conformal prediction can be used to fine-tune the balance between hit retrieval and hit structural diversity in a virtual screening setting. PCM can be applied to future virtual screening and compound design, including off-target prediction for bromodomains.

Dimeric calpains constitute a promising therapeutic target for many diseases such as cardiovascular, neurodegenerative and ischaemic disease. The discovery of selective calpain inhibitors, however, has been extremely challenging. Previously, allosteric inhibitors of calpains, such as PD150606, which included a specific α-mercaptoacrylic acid sub-structure, were reported to bind to the penta-EF hand calcium binding domain, PEF(S) of calpain. Although these are selective to calpains over other cysteine proteases, their mode of action has remained elusive due to their ability to inhibit the active site domain with and without the presence of PEF(S), with similar potency. These findings have led to the question of whether the inhibitory response can be attributed to an allosteric mode of action or alternatively to inhibition at the active site. In order to address this problem, we report a structure-based virtual screening protocol as a novel approach for the discovery of PEF(S) binders that populate a novel chemical space. We have identified compound 1, Vidupiprant, which is shown to bind to the PEF(S) domain by the TNS displacement method, and it exhibited specificity in its allosteric mode of inhibition. Compound 1 inhibited the full-length calpain-1 complex with a higher potency (IC50 = 7.5 μM) than the selective, cell-permeable non-peptide calpain inhibitor, PD150606 (IC50 = 19.3 μM), where the latter also inhibited the active site domain in the absence of PEF(S) (IC50 = 17.8 μM). Hence the method presented here has identified known compounds with a novel allosteric mechanism for the inhibition of calpain-1. We show for the first time that the inhibition of enzyme activity can be attributed to an allosteric mode of action, which may offer improved selectivity and a reduced side-effects profile.

The versatility of similarity searching and quantitative structure-activity relationships to model the activity of compound sets within given bioactivity ranges (i.e., interpolation) is well established. However, their relative performance in the common scenario in early stage drug discovery where lots of inactive data but no active data points are available (i.e., extrapolation from the low-activity to the high-activity range) has not been thoroughly examined yet. To this aim, we have designed an iterative virtual screening strategy which was evaluated on 25 diverse bioactivity data sets from ChEMBL. We benchmark the efficiency of random forest (RF), multiple linear regression, ridge regression, similarity searching, and random selection of compounds to identify a highly active molecule in the test set among a large number of low-potency compounds. We use the number of iterations required to find this active molecule to evaluate the performance of each experimental setup. We show that linear and ridge regression often outperform RF and similarity searching, reducing the number of iterations to find an active compound by a factor of 2 or more. Even simple regression methods seem better able to extrapolate to high-bioactivity ranges than RF, which only provides output values in the range covered by the training set. In addition, examination of the scaffold diversity in the data sets used shows that in some cases similarity searching and RF require two times as many iterations as random selection depending on the chemical space covered in the initial training data. Lastly, we show using bioactivity data for COX-1 and COX-2 that our framework can be extended to multitarget drug discovery, where compounds are selected by concomitantly considering their activity against multiple targets. Overall, this study provides an approach for iterative screening where only inactive data are present in early stages of drug discovery in order to discover highly potent compounds and the best experimental set up in which to do so.

This review highlights developing methods in the toxicogenomics field and their applications to understanding and predicting compound induced toxicity., The toxicogenomics field aims to understand and predict toxicity by using ‘omics’ data in order to study systems-level responses to compound treatments. In recent years there has been a rapid increase in publicly available toxicological and ‘omics’ data, particularly gene expression data, and a corresponding development of methods for its analysis. In this review, we summarize recent progress relating to the analysis of RNA-Seq and microarray data, review relevant databases, and highlight recent applications of toxicogenomics data for understanding and predicting compound toxicity. These include the analysis of differentially expressed genes and their enrichment, signature matching, methods based on interaction networks, and the analysis of co-expression networks. In the future, these state-of-the-art methods will likely be combined with new technologies, such as whole human body models, to produce a comprehensive systems-level understanding of toxicity that reduces the necessity of in vivo toxicity assessment in animal models.

Background Nearly half of the world’s population (3.2 billion people) were at risk of malaria in 2015, and resistance to current therapies is a major concern. While the standard of care includes drug combinations, there is a pressing need to identify new combinations that can bypass current resistance mechanisms. In the work presented here, a combined transcriptional drug repositioning/discovery and machine learning approach is proposed. Methods The integrated approach utilizes gene expression data from patient-derived samples, in combination with large-scale anti-malarial combination screening data, to predict synergistic compound combinations for three Plasmodium falciparum strains (3D7, DD2 and HB3). Both single compounds and combinations predicted to be active were prospectively tested in experiment. Results One of the predicted single agents, apicidin, was active with the AC50 values of 74.9, 84.1 and 74.9 nM in 3D7, DD2 and HB3 P. falciparum strains while its maximal safe plasma concentration in human is 547.6 ± 136.6 nM. Apicidin at the safe dose of 500 nM kills on average 97% of the parasite. The synergy prediction algorithm exhibited overall precision and recall of 83.5 and 65.1% for mild-to-strong, 48.8 and 75.5% for moderate-to-strong and 12.0 and 62.7% for strong synergies. Some of the prospectively predicted combinations, such as tacrolimus-hydroxyzine and raloxifene-thioridazine, exhibited significant synergy across the three P. falciparum strains included in the study. Conclusions Systematic approaches can play an important role in accelerating discovering novel combinational therapies for malaria as it enables selecting novel synergistic compound pairs in a more informed and cost-effective manner. Electronic supplementary material The online version of this article (10.1186/s12936-018-2294-5) contains supplementary material, which is available to authorized users.

Phosphodiesterases are proving to be fruitful targets for drug discovery. At the same time fragment-based drug discovery has matured into a powerful and widely applied technique. In this communication we review the application of fragment-based drug discovery for the successful identification of novel 3',5'-cyclic nucleotide phosphodiesterase (PDE) inhibitors, concentrating on both experimental and computational strategies for fragment screening and hit-to-lead development. To this end, we also mine the open access databases ChEMBL and PDB for fragments showing PDE inhibitory activity, as well as SureChEMBL for recent PDE related patents, to provide a wider context for exploring fragment diversity. Together these approaches form an integrated experimental and computational platform to exploit fragment-based drug discovery for this important gene superfamily.

Iterative screening has emerged as a promising approach to increase the efficiency of screening campaigns compared to traditional high throughput approaches. By learning from a subset of the compound library, inferences on what compounds to screen next can be made by predictive models, resulting in more efficient screening. One way to evaluate screening is to consider the cost of screening compared to the gain associated with finding an active compound. In this work, we introduce a conformal predictor coupled with a gain-cost function with the aim to maximise gain in iterative screening. Using this setup we were able to show that by evaluating the predictions on the training data, very accurate predictions on what settings will produce the highest gain on the test data can be made. We evaluate the approach on 12 bioactivity datasets from PubChem training the models using 20% of the data. Depending on the settings of the gain-cost function, the settings generating the maximum gain were accurately identified in 8–10 out of the 12 datasets. Broadly, our approach can predict what strategy generates the highest gain based on the results of the cost-gain evaluation: to screen the compounds predicted to be active, to screen all the remaining data, or not to screen any additional compounds. When the algorithm indicates that the predicted active compounds should be screened, our approach also indicates what confidence level to apply in order to maximize gain. Hence, our approach facilitates decision-making and allocation of the resources where they deliver the most value by indicating in advance the likely outcome of a screening campaign., The research at Swetox (UN) was supported by Knut and Alice Wallenberg Foundation and Swedish Research Council FORMAS. AMA was supported by AstraZeneca.

Activity landscape modeling is mostly a descriptive technique that allows rationalizing continuous and discontinuous SARs. Nevertheless, the interpretation of some landscape features, especially of activity cliffs, is not straightforward. As the nature of activity cliffs depends on the ligand and the target, information regarding both should be included in the analysis. A specific way to include this information is using protein-ligand interaction fingerprints (IFPs). In this paper we report the activity landscape modeling of 507 ligand-kinase complexes (from the KLIFS database) including IFP, which facilitates the analysis and interpretation of activity cliffs. Here we introduce the structure-activity-interaction similarity (SAIS) maps that incorporate information on ligand-target contact similarity. We also introduce the concept of interaction cliffs defined as ligand-target complexes with high structural and interaction similarity but have a large potency difference of the ligands. Moreover, the information retrieved regarding the specific interaction allowed the identification of activity cliff hot spots, which help to rationalize activity cliffs from the target point of view. In general, the information provided by IFPs provides a structure-based understanding of some activity landscape features. This paper shows examples of analyses that can be carried out when IFPs are added to the activity landscape model.

In silico protein target deconvolution is frequently used for mechanism-of-action investigations; however existing protocols usually do not predict compound functional effects, such as activation or inhibition, upon binding to their protein counterparts. This study is hence concerned with including functional effects in target prediction. To this end, we assimilated a bioactivity training set for 332 targets, comprising 817,239 active data points with unknown functional effect (binding data) and 20,761,260 inactive compounds, along with 226,045 activating and 1,032,439 inhibiting data points from functional screens. Chemical space analysis of the data first showed some separation between compound sets (binding and inhibiting compounds were more similar to each other than both binding and activating or activating and inhibiting compounds), providing a rationale for implementing functional prediction models. We employed three different architectures to predict functional response, ranging from simplistic random forest models ('Arch1') to cascaded models which use separate binding and functional effect classification steps ('Arch2' and 'Arch3'), differing in the way training sets were generated. Fivefold stratified cross-validation outlined cascading predictions provides superior precision and recall based on an internal test set. We next prospectively validated the architectures using a temporal set of 153,467 of in-house data points (after a 4-month interim from initial data extraction). Results outlined Arch3 performed with the highest target class averaged precision and recall scores of 71% and 53%, which we attribute to the use of inactive background sets. Distance-based applicability domain (AD) analysis outlined that Arch3 provides superior extrapolation into novel areas of chemical space, and thus based on the results presented here, propose as the most suitable architecture for the functional effect prediction of small molecules. We finally conclude including functional effects could provide vital insight in future studies, to annotate cases of unanticipated functional changeover, as outlined by our CHRM1 case study.

Compounds designed to display polypharmacology may have utility in treating complex diseases, where activity at multiple targets is required to produce a clinical efect. In particular, suitable compounds may be useful in treating neurodegenerative diseases by promoting neuronal survival in a synergistic manner via their multi-target activity at the adenosine A1 and A2A receptors (A1R and A2AR) and phosphodiesterase 10A (PDE10A), which modulate intracellular cAMP levels. Hence, in this work we describe a computational method for the design of synthetically feasible ligands that bind to A1 and A2A receptors and inhibit phosphodiesterase 10A (PDE10A), involving a retrosynthetic approach employing in silico target prediction and docking, which may be generally applicable to multi-target compound design at several target classes. This approach has identifed 2-aminopyridine-3-carbonitriles as the frst multi-target ligands at A1R, A2AR and PDE10A, by showing agreement between the ligand and structure based predictions at these targets. The series were synthesized via an efcient one-pot scheme and validated pharmacologically as A1R/A2AR–PDE10A ligands, with IC50 values of 2.4–10.0 μM at PDE10A and Ki values of 34–294 nM at A1R and/or A2AR. Furthermore, selectivity profling of the synthesized 2-amino-pyridin-3-carbonitriles against other subtypes of both protein families showed that the multi-target ligand 8 exhibited a minimum of twofold selectivity over all tested oftargets. In addition, both compounds 8 and 16 exhibited the desired multi-target profle, which could be considered for further functional efcacy assessment, analog modifcation for the improvement of selectivity towards A1R, A2AR and PDE10A collectively, and evaluation of their potential synergy in modulating cAMP levels LK thanks the IDB Cambridge International Scholarship for support. This work was fnancially supported by the ERC Starting Grant to AB (No. 336159), the Consellería de Cultura, Educación e Ordenación Universitaria of the Galician Government: (Grant: GPC2014/03), Centro singular de investigación de Galicia accreditation 2016–2019 (ED431G/09) and the European Regional Development Fund (ERDF) SI

It is well established that the number of publications of novel small molecule drugs, and their associated targets, has increased over the years. This work provides an update on publishing trends over the years with a particular focus on the comparison between patents and scientific literature of which are accessible through ChEMBL and GOSTAR databases. More precisely, the patents and scientific literature associated with bioactive molecules and their target annotations have been compared to identify where novelty originated from. To analyse potential target class influences, the data has been further split into eight different target classes.. Moreover, small molecule modulators for protein targets are usually published in both scientific literature and in patents (45%), or only in scientific literature (51%) but rarely in patents only. It has been observed that generally, novel targets and their associated compounds are published in literature primarily, whereas novel compounds (regardless of their associated targets) tend to be published in patents first., BBSRC and AstraZeneca

In this work, we describe the computational ("in silico") mode-of-action analysis of CNS-active drugs, which is taking both multiple simultaneous hypotheses as well as sets of protein targets for each mode-of-action into account, and which was followed by successful prospective in vitro and in vivo validation. Using sleep-related phenotypic readouts describing both efficacy and side effects for 491 compounds tested in rat, we defined an "optimal" (desirable) sleeping pattern. Compounds were subjected to in silico target prediction (which was experimentally confirmed for 21 out of 28 cases), followed by the utilization of decision trees for deriving polypharmacological bioactivity profiles. We demonstrated that predicted bioactivities improved classification performance compared to using only structural information. Moreover, DrugBank molecules were processed via the same pipeline, and compounds in many cases not annotated as sedative-hypnotic (alcaftadine, benzatropine, palonosetron, ecopipam, cyproheptadine, sertindole, and clopenthixol) were prospectively validated in vivo. Alcaftadine, ecopipam cyproheptadine, and clopenthixol were found to promote sleep as predicted, benzatropine showed only a small increase in NREM sleep, whereas sertindole promoted wakefulness. To our knowledge, the sedative-hypnotic effects of alcaftadine and ecopipam have not been previously discussed in the literature. The method described extends previous single-target, single-mode-of-action models and is applicable across disease areas.

High-throughput screening, where thousands of molecules rapidly can be assessed for activity against a protein, has been the dominating approach in drug discovery for many years. However, these methods are costly and require much time and effort. In order to suggest an improvement to this situation, in this study, we apply an iterative screening process, where an initial set of compounds are selected for screening based on molecular docking. The outcome of the initial screen is then used to classify the remaining compounds through a conformal predictor. The approach was retrospectively validated using 41 targets from the Directory of Useful Decoys, Enhanced (DUD-E), ensuring scaffold diversity among the active compounds. The results show that 57% of the remaining active compounds could be identified while only screening 9.4% of the database. The overall hit rate (7.6%) was also higher than when using docking alone (5.2%). When limiting the search to the top scored compounds from docking, 39.6% of the active compounds could be identified, compared to 13.5% when screening the same number of compounds solely based on docking. The use of conformal predictors also gives a clear indication of the number of compounds to screen in the next iteration. These results indicate that iterative screening based on molecular docking and conformal prediction can be an efficient way to find active compounds while screening only a small part of the compound collection.

One important, however, poorly understood, concept of Traditional Chinese Medicine (TCM) is that of hot, cold, and neutral nature of its bioactive principles. To advance the field, in this study, we analyzed compound-nature pairs from TCM on a large scale (>23 000 structures) via chemical space visualizations to understand its physicochemical domain and in silico target prediction to understand differences related to their modes-of-action (MoA) against proteins. We found that overall TCM natures spread into different subclusters with specific molecular patterns, as opposed to forming coherent global groups. Compounds associated with cold nature had a lower clogP and contain more aliphatic rings than the other groups and were found to control detoxification, heat-clearing, heart development processes, and have sedative function, associated with "Mental and behavioural disorders" diseases. While compounds associated with hot nature were on average of lower molecular weight, have more aromatic ring systems than other groups, frequently seemed to control body temperature, have cardio-protection function, improve fertility and sexual function, and represent excitatory or activating effects, associated with "endocrine, nutritional and metabolic diseases" and "diseases of the circulatory system". Compounds associated with neutral nature had a higher polar surface area and contain more cyclohexene moieties than other groups and seem to be related to memory function, suggesting that their nature may be a useful guide for their utility in neural degenerative diseases. We were hence able to elucidate the difference between different nature classes in TCM on the molecular level, and on a large data set, for the first time, thereby helping a better understanding of TCM nature theory and bridging the gap between traditional medicine and our current understanding of the human body.

Background Expression and activity of heparanase, an endoglycosidase that cleaves heparan sulfate (HS) side chains of proteoglycans, is associated with progression and poor prognosis of many cancers which makes it an attractive drug target in cancer therapeutics. Methods In the present work, we report the in vitro screening of a library of 150 small molecules with the scaffold bearing quinolones, oxazines, benzoxazines, isoxazoli(di)nes, pyrimidinones, quinolines, benzoxazines, and 4-thiazolidinones, thiadiazolo[3,2-a]pyrimidin-5-one, 1,2,4-triazolo-1,3,4-thiadiazoles, and azaspiranes against the enzymatic activity of human heparanase. The identified lead compounds were evaluated for their heparanase-inhibiting activity using sulfate [35S] labeled extracellular matrix (ECM) deposited by cultured endothelial cells. Further, anti-invasive efficacy of lead compound was evaluated against hepatocellular carcinoma (HepG2) and Lewis lung carcinoma (LLC) cells. Results Among the 150 compounds screened, we identified 1,2,4-triazolo-1,3,4-thiadiazoles bearing compounds to possess human heparanase inhibitory activity. Further analysis revealed 2,4-Diiodo-6-(3-phenyl-[1, 2, 4]triazolo[3,4-b][1, 3, 4]thiadiazol-6yl)phenol (DTP) as the most potent inhibitor of heparanase enzymatic activity among the tested compounds. The inhibitory efficacy was demonstrated by a colorimetric assay and further validated by measuring the release of radioactive heparan sulfate degradation fragments from [35S] labeled extracellular matrix. Additionally, lead compound significantly suppressed migration and invasion of LLC and HepG2 cells with IC50 value of ~5 μM. Furthermore, molecular docking analysis revealed a favourable interaction of triazolo-thiadiazole backbone with Asn-224 and Asp-62 of the enzyme. Conclusions Overall, we identified biologically active heparanase inhibitor which could serve as a lead structure in developing compounds that target heparanase in cancer. Electronic supplementary material The online version of this article (doi:10.1186/s12885-017-3214-8) contains supplementary material, which is available to authorized users.

Similarity in the ligand binding profile of two enzymes may provide insight for functional characterization and be of greater relevance to inhibitor development than sequence similarity or structural homology. Fragment screening is an efficient approach to characterizing the ligand profile of an enzyme and has been applied here to study the family of cytochrome P450 enzymes (P450s) expressed by Mycobacterium tuberculosis (Mtb). The Mtb P450s have important roles in bacterial virulence, survival and pathogenicity. Comparing the fragment profiles of seven of these enzymes revealed that P450s which share a similar biological function have significantly similar fragment profiles, while functionally unrelated or orphan P450s exhibit distinct ligand binding properties, despite overall high structural homology. Chemical structures that exhibit promiscuous binding between enzymes have been identified, as have selective fragments that could provide leads for inhibitor development. The similarity in the fragment binding profile of the orphan enzyme CYP144A1 to CYP121A1, an enzyme important for Mtb viability, provides a case study illustrating the subsequent identification of novel CYP144A1 ligands. The different binding modes of these compounds to CYP144A1 provide insight into structural and dynamic aspects of the enzyme, suggest a hypothesis into biological function and provide opportunity for inhibitor development. Expanding this fragment profiling approach to include a greater number of functionally characterized and orphan proteins may provide a valuable resource for understanding enzyme-ligand interactions.

International audience; Serine proteases, implicated in important physiological functions, have a high intra-family similarity, which leads to unwanted off-target effects of inhibitors with insufficient selectivity. However, the availability of sequence and structure data has now made it possible to develop approaches to design pharmacological agents that can discriminate successfully between their related binding sites. In this study, we have quantified the relationship between 12 625 distinct protease inhibitors and their bioactivity against 67 targets of the serine protease family (20 213 data points) in an integrative manner, using proteochemometric modelling (PCM). The benchmarking of 21 different target descriptors motivated the usage of specific binding pocket amino acid descriptors, which helped in the identification of active site residues and selective compound chemotypes affecting compound affinity and selectivity. PCM models performed better than alternative approaches (models trained using exclusively compound descriptors on all available data, QSAR) employed for comparison with R 2 /RMSE values of 0.64 AE 0.23/ 0.66 AE 0.20 vs. 0.35 AE 0.27/1.05 AE 0.27 log units, respectively. Moreover, the interpretation of the PCM model singled out various chemical substructures responsible for bioactivity and selectivity towards particular proteases (thrombin, trypsin and coagulation factor 10) in agreement with the literature. For instance, absence of a tertiary sulphonamide was identified to be responsible for decreased selective activity (by on average 0.27 AE 0.65 pChEMBL units) on FA10. Among the binding pocket residues, the amino acids (arginine, leucine and tyrosine) at positions 35, 39, 60, 93, 140 and 207 were observed as key contributing residues for selective affinity on these three targets.

The assessment of compound cytotoxicity is an important part of the drug discovery process. Accurate predictions of cytotoxicity have the potential to expedite decision making and save considerable time and effort. In this work we apply class conditional conformal prediction to model the cytotoxicity of compounds based on 16 high throughput cytotoxicity assays from PubChem. The data span 16 cell lines and comprise more than 440 000 unique compounds. The data sets are heavily imbalanced with only 0.8% of the tested compounds being cytotoxic. We trained one classification model for each cell line and validated the performance with respect to validity and accuracy. The generated models deliver high quality predictions for both toxic and non-toxic compounds despite the imbalance between the two classes. On external data collected from the same assay provider as one of the investigated cell lines the model had a sensitivity of 74% and a specificity of 65% at the 80% confidence level among the compounds assigned to a single class. Compared to previous approaches for large scale cytotoxicity modelling, this represents a balanced performance in the prediction of the toxic and non-toxic classes. The conformal prediction framework also allows the modeller to control the error frequency of the predictions, allowing predictions of cytotoxicity outcomes with confidence.

Combination antibiotic therapies are clinically important in the fight against bacterial infections. However, the search space of drug combinations is large, making the identification of effective combinations a challenging task. Here, we present a computational framework that uses substructure profiles derived from the molecular structures of drugs and predicts antibiotic interactions. Using a previously published data set of 153 drug pairs, we showed that substructure profiles are useful in predicting synergy. We experimentally measured the interaction of 123 new drug pairs, as a prospective validation set for our approach, and identified 37 new synergistic pairs. Of the 12 pairs predicted to be synergistic, 10 were experimentally validated, corresponding to a 2.8-fold enrichment. Having thus validated our methodology, we produced a compendium of interaction predictions for all pairwise combinations among 100 antibiotics. Our methodology can make reliable antibiotic interaction predictions for any antibiotic pair within the applicability domain of the model since it solely requires chemical structures as an input.

While mechanisms of cytotoxicity and cytostaticity have been studied extensively from the biological side, relatively little is currently understood regarding areas of chemical space leading to cytotoxicity and cytostasis in large compound collections. Predicting and rationalizing potential adverse mechanism-of-actions (MoAs) of small molecules is however crucial for screening library design, given the link of even low level cytotoxicity and adverse events observed in man. In this study, we analyzed results from a cell-based cytotoxicity screening cascade, comprising 296 970 nontoxic, 5784 cytotoxic and cytostatic, and 2327 cytostatic-only compounds evaluated on the THP-1 cell-line. We employed an in silico MoA analysis protocol, utilizing 9.5 million active and 602 million inactive bioactivity points to generate target predictions, annotate predicted targets with pathways, and calculate enrichment metrics to highlight targets and pathways. Predictions identify known mechanisms for the top ranking targets and pathways for both phenotypes after review and indicate that while processes involved in cytotoxicity versus cytostaticity seem to overlap, differences between both phenotypes seem to exist to some extent. Cytotoxic predictions highlight many kinases, including the potentially novel cytotoxicity-related target STK32C, while cytostatic predictions outline targets linked with response to DNA damage, metabolism, and cytoskeletal machinery. Fragment analysis was also employed to generate a library of toxicophores to improve general understanding of the chemical features driving toxicity. We highlight substructures with potential kinase-dependent and kinase-independent mechanisms of toxicity. We also trained a cytotoxic classification model on proprietary and public compound readouts, and prospectively validated these on 988 novel compounds comprising difficult and trivial testing instances, to establish the applicability domain of models. The proprietary model performed with precision and recall scores of 77.9% and 83.8%, respectively. The MoA results and top ranking substructures with accompanying MoA predictions are available as a platform to assess screening collections.

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($\alpha$ = 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

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.

Small molecules are being increasingly used for inducing the targeted differentiation of stem cells to different cell types. However, until now no systematic method for selecting suitable small molecules for this purpose has been presented. In this work, we propose an integrated and general bioinformatics- and cheminformatics-based approach for selecting small molecules which direct cellular differentiation in the desired way. The approach was successfully experimentally validated for differentiating stem cells into cardiomyocytes. All predicted compounds enhanced expression of cardiac progenitor (Gata4, Nkx2-5 and Mef2c) and mature cardiac markers (Actc1, myh6) significantly during and post-cardiac progenitor formation. The best-performing compound, Famotidine, increased the percentage of Myh6-positive cells from 33 to 56%, and enhanced the expression of Nkx2.5 and Tnnt2 cardiac progenitor and cardiac markers in protein level. The approach employed in the study is applicable to all other stem cell differentiation settings where gene expression data are available.

Metabolism of xenobiotics remains a central challenge for the discovery and development of drugs, cosmetics, nutritional supplements, and agrochemicals. Metabolic transformations are frequently related to the incidence of toxic effects that may result from the emergence of reactive species, the systemic accumulation of metabolites, or by induction of metabolic pathways. Experimental investigation of the metabolism of small organic molecules is particularly resource demanding; hence, computational methods are of considerable interest to complement experimental approaches. This review provides a broad overview of structure- and ligand-based computational methods for the prediction of xenobiotic metabolism. Current computational approaches to address xenobiotic metabolism are discussed from three major perspectives: (i) prediction of sites of metabolism (SOMs), (ii) elucidation of potential metabolites and their chemical structures, and (iii) prediction of direct and indirect effects of xenobiotics on metabolizing enzymes, where the focus is on the cytochrome P450 (CYP) superfamily of enzymes, the cardinal xenobiotics metabolizing enzymes. For each of these domains, a variety of approaches and their applications are systematically reviewed, including expert systems, data mining approaches, quantitative structure-activity relationships (QSARs), and machine learning-based methods, pharmacophore-based algorithms, shape-focused techniques, molecular interaction fields (MIFs), reactivity-focused techniques, protein-ligand docking, molecular dynamics (MD) simulations, and combinations of methods. Predictive metabolism is a developing area, and there is still enormous potential for improvement. However, it is clear that the combination of rapidly increasing amounts of available ligand- and structure-related experimental data (in particular, quantitative data) with novel and diverse simulation and modeling approaches is accelerating the development of effective tools for prediction of in vivo metabolism, which is reflected by the diverse and comprehensive data sources and methods for metabolism prediction reviewed here. This review attempts to survey the range and scope of computational methods applied to metabolism prediction and also to compare and contrast their applicability and performance.

Traditional Chinese medicine (TCM) still needs more scientific rationale to be proven for it to be accepted further in the West. We are now in the position to propose computational hypotheses for the mode-of-actions (MOAs) of 45 TCM therapeutic action (sub)classes fromin silicotarget prediction algorithms, whose target was later annotated with Kyoto Encyclopedia of Genes and Genomes pathway, and to discover the relationship between them by generating a hierarchical clustering. The results of 10,749 TCM compounds showed 183 enriched targets and 99 enriched pathways from Estimation Score ≤ 0 and ≥ 5% of compounds/targets in a (sub)class. The MOA of a (sub)class was established from supporting literature. Overall, the most frequent top three enriched targets/pathways were immune-related targets such as tyrosine-protein phosphatase nonreceptor type 2 (PTPN2) and digestive system such as mineral absorption. We found two major protein families, G-protein coupled receptor (GPCR), and protein kinase family contributed to the diversity of the bioactivity space, while digestive system was consistently annotated pathway motif, which agreed with the important treatment principle of TCM, “the foundation of acquired constitution” that includes spleen and stomach. In short, the TCM (sub)classes, in many cases share similar targets/pathways despite having different indications.

Background In silico predictive models have proved to be valuable for the optimisation of compound potency, selectivity and safety profiles in the drug discovery process. Results camb is an R package that provides an environment for the rapid generation of quantitative Structure-Property and Structure-Activity models for small molecules (including QSAR, QSPR, QSAM, PCM) and is aimed at both advanced and beginner R users. camb's capabilities include the standardisation of chemical structure representation, computation of 905 one-dimensional and 14 fingerprint type descriptors for small molecules, 8 types of amino acid descriptors, 13 whole protein sequence descriptors, filtering methods for feature selection, generation of predictive models (using an interface to the R package caret), as well as techniques to create model ensembles using techniques from the R package caretEnsemble). Results can be visualised through high-quality, customisable plots (R package ggplot2). Conclusions Overall, camb constitutes an open-source framework to perform the following steps: (1) compound standardisation, (2) molecular and protein descriptor calculation, (3) descriptor pre-processing and model training, visualisation and validation, and (4) bioactivity/property prediction for new molecules. camb aims to speed model generation, in order to provide reproducibility and tests of robustness. QSPR and proteochemometric case studies are included which demonstrate camb's application.Graphical abstractFrom compounds and data to models: a complete model building workflow in one package. Electronic supplementary material The online version of this article (doi:10.1186/s13321-015-0086-2) contains supplementary material, which is available to authorized users.

Background In silico analyses are increasingly being used to support mode-of-action investigations; however many such approaches do not utilise the large amounts of inactive data held in chemogenomic repositories. The objective of this work is concerned with the integration of such bioactivity data in the target prediction of orphan compounds to produce the probability of activity and inactivity for a range of targets. To this end, a novel human bioactivity data set was constructed through the assimilation of over 195 million bioactivity data points deposited in the ChEMBL and PubChem repositories, and the subsequent application of a sphere-exclusion selection algorithm to oversample presumed inactive compounds. Results A Bernoulli Naïve Bayes algorithm was trained using the data and evaluated using fivefold cross-validation, achieving a mean recall and precision of 67.7 and 63.8 % for active compounds and 99.6 and 99.7 % for inactive compounds, respectively. We show the performances of the models are considerably influenced by the underlying intraclass training similarity, the size of a given class of compounds, and the degree of additional oversampling. The method was also validated using compounds extracted from WOMBAT producing average precision-recall AUC and BEDROC scores of 0.56 and 0.85, respectively. Inactive data points used for this test are based on presumed inactivity, producing an approximated indication of the true extrapolative ability of the models. A distance-based applicability domain analysis was also conducted; indicating an average Tanimoto Coefficient distance of 0.3 or greater between a test and training set can be used to give a global measure of confidence in model predictions. A final comparison to a method trained solely on active data from ChEMBL performed with precision-recall AUC and BEDROC scores of 0.45 and 0.76. Conclusions The inclusion of inactive data for model training produces models with superior AUC and improved early recognition capabilities, although the results from internal and external validation of the models show differing performance between the breadth of models. The realised target prediction protocol is available at https://github.com/lhm30/PIDGIN.Graphical abstractThe inclusion of large scale negative training data for in silico target prediction improves the precision and recall AUC and BEDROC scores for target models. Electronic supplementary material The online version of this article (doi:10.1186/s13321-015-0098-y) contains supplementary material, which is available to authorized users.