Search

Background and Objective: Clinical pharmacological modeling and statistical analysis software is an essential basic tool for drug development and personalized drug therapy. The learning curve of current basic tools is steep and unfriendly to beginners. The curve is even more challenging in cases of significant individual differences or measurement errors in data, resulting in difficulties in accurately estimating pharmacokinetic parameters by existing fitting algorithms. Hence, this study aims to explore a new optimized parameter fitting algorithm that reduces the sensitivity of the model to initial values and integrate it into the CPhaMAS platform, a user-friendly online application for pharmacokinetic data analysis., Methods: In this study, we proposed an optimized Nelder-Mead method that reinitializes simplex vertices when trapped in local solutions and integrated it into the CPhaMAS platform. The CPhaMAS, an online platform for pharmacokinetic data analysis, includes three modules: compartment model analysis, non-compartment analysis (NCA) and bioequivalence/bioavailability (BE/BA) analysis. Our proposed CPhaMAS platform was evaluated and compared with existing WinNonlin., Results: The platform was easy to learn and did not require code programming. The accuracy investigation found that the optimized Nelder-Mead method of the CPhaMAS platform showed better accuracy (smaller mean relative error and higher R 2 ) in two-compartment and extravascular administration models when the initial value was set to true and abnormal values (10 times larger or smaller than the true value) compared with the WinNonlin. The mean relative error of the NCA calculation parameters of CPhaMAS and WinNonlin was <0.0001 %. When calculating BE for conventional, high-variability and narrow-therapeutic drugs. The main statistical parameters of the parameters C max , AUC t , and AUC inf in CPhaMAS have a mean relative error of <0.01% compared to WinNonLin., Conclusions: In summary, CPhaMAS is a user-friendly platform with relatively accurate algorithms. It is a powerful tool for analysing pharmacokinetic data for new drug development and precision medicine., Competing Interests: Declaration of competing interest The authors declared no competing interests for this work., (Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.)

Liver microsomal stability, a crucial aspect of metabolic stability, significantly impacts practical drug discovery. However, current models for predicting liver microsomal stability are based on limited molecular information from a single species. To address this limitation, we constructed the largest public database of compounds from three common species: human, rat, and mouse. Subsequently, we developed a series of classification models using both traditional descriptor-based and classic graph-based machine learning (ML) algorithms. Remarkably, the best-performing models for the three species achieved Matthews correlation coefficients (MCCs) of 0.616, 0.603, and 0.574, respectively, on the test set. Furthermore, through the construction of consensus models based on these individual models, we have demonstrated their superior predictive performance in comparison with the existing models of the same type. To explore the similarities and differences in the properties of liver microsomal stability among multispecies molecules, we conducted preliminary interpretative explorations using the Shapley additive explanations (SHAP) and atom heatmap approaches for the models and misclassified molecules. Additionally, we further investigated representative structural modifications and substructures that decrease the liver microsomal stability in different species using the matched molecule pair analysis (MMPA) method and substructure extraction techniques. The established prediction models, along with insightful interpretation information regarding liver microsomal stability, will significantly contribute to enhancing the efficiency of exploring practical drugs for development.

Patents play a crucial role in drug research and development, providing early access to unpublished data and offering unique insights. Identifying key compounds in patents is essential to finding novel lead compounds. This study collected a comprehensive data set comprising 1555 patents, encompassing 1000 key compounds, to explore innovative approaches for predicting these key compounds. Our novel PatentNetML framework integrated network science and machine learning algorithms, combining network measures, ADMET properties, and physicochemical properties, to construct robust classification models to identify key compounds. Through a model interpretation and an analysis of three compelling case studies, we showcase the potential of PatentNetML in unveiling hidden patterns and connections within diverse patents. While our framework is pioneering, we acknowledge its limitations when applied to patents that deviate from the assumed central pattern. This work serves as a promising foundation for future research endeavors aimed at efficiently identifying promising drug candidates and expediting drug discovery in the pharmaceutical industry.

Drug discovery and development constitute a laborious and costly undertaking. The success of a drug hinges not only good efficacy but also acceptable absorption, distribution, metabolism, elimination, and toxicity (ADMET) properties. Overall, up to 50% of drug development failures have been contributed from undesirable ADMET profiles. As a multiple parameter objective, the optimization of the ADMET properties is extremely challenging owing to the vast chemical space and limited human expert knowledge. In this study, a freely available platform called Chemical Molecular Optimization, Representation and Translation (ChemMORT) is developed for the optimization of multiple ADMET endpoints without the loss of potency (https://cadd.nscc-tj.cn/deploy/chemmort/). ChemMORT contains three modules: Simplified Molecular Input Line Entry System (SMILES) Encoder, Descriptor Decoder and Molecular Optimizer. The SMILES Encoder can generate the molecular representation with a 512-dimensional vector, and the Descriptor Decoder is able to translate the above representation to the corresponding molecular structure with high accuracy. Based on reversible molecular representation and particle swarm optimization strategy, the Molecular Optimizer can be used to effectively optimize undesirable ADMET properties without the loss of bioactivity, which essentially accomplishes the design of inverse QSAR. The constrained multi-objective optimization of the poly (ADP-ribose) polymerase-1 inhibitor is provided as the case to explore the utility of ChemMORT., (© The Author(s) 2024. Published by Oxford University Press.)

Detecting drug-drug interactions (DDIs) is an essential step in drug development and drug administration. Given the shortcomings of current experimental methods, the machine learning (ML) approach has become a reliable alternative, attracting extensive attention from the academic and industrial fields. With the rapid development of computational science and the growing popularity of cross-disciplinary research, a large number of DDI prediction studies based on ML methods have been published in recent years. To give an insight into the current situation and future direction of DDI prediction research, we systemically review these studies from three aspects: (1) the classic DDI databases, mainly including databases of drugs, side effects, and DDI information; (2) commonly used drug attributes, which focus on chemical, biological, and phenotypic attributes for representing drugs; (3) popular ML approaches, such as shallow learning-based, deep learning-based, recommender system-based, and knowledge graph-based methods for DDI detection. For each section, related studies are described, summarized, and compared, respectively. In the end, we conclude the research status of DDI prediction based on ML methods and point out the existing issues, future challenges, potential opportunities, and subsequent research direction.

Motivation: Spatial clustering is essential and challenging for spatial transcriptomics' data analysis to unravel tissue microenvironment and biological function. Graph neural networks are promising to address gene expression profiles and spatial location information in spatial transcriptomics to generate latent representations. However, choosing an appropriate graph deep learning module and graph neural network necessitates further exploration and investigation., Results: In this article, we present GRAPHDeep to assemble a spatial clustering framework for heterogeneous spatial transcriptomics data. Through integrating 2 graph deep learning modules and 20 graph neural networks, the most appropriate combination is decided for each dataset. The constructed spatial clustering method is compared with state-of-the-art algorithms to demonstrate its effectiveness and superiority. The significant new findings include: (i) the number of genes or proteins of spatial omics data is quite crucial in spatial clustering algorithms; (ii) the variational graph autoencoder is more suitable for spatial clustering tasks than deep graph infomax module; (iii) UniMP, SAGE, SuperGAT, GATv2, GCN, and TAG are the recommended graph neural networks for spatial clustering tasks; and (iv) the used graph neural network in the existent spatial clustering frameworks is not the best candidate. This study could be regarded as desirable guidance for choosing an appropriate graph neural network for spatial clustering., Availability and Implementation: The source code of GRAPHDeep is available at https://github.com/narutoten520/GRAPHDeep. The studied spatial omics data are available at https://zenodo.org/record/8141084., (© The Author(s) 2024. Published by Oxford University Press.)

Adverse drug events (ADEs) are common in clinical practice and can cause significant harm to patients and increase resource use. Natural language processing (NLP) has been applied to automate ADE detection, but NLP systems become less adaptable when drug entities are missing or multiple medications are specified in clinical narratives. Additionally, no Chinese-language NLP system has been developed for ADE detection due to the complexity of Chinese semantics, despite ˃10 million cases of drug-related adverse events occurring annually in China. To address these challenges, we propose DKADE, a deep learning and knowledge graph-based framework for identifying ADEs. DKADE infers missing drug entities and evaluates their correlations with ADEs by combining medication orders and existing drug knowledge. Moreover, DKADE can automatically screen for new adverse drug reactions. Experimental results show that DKADE achieves an overall F1-score value of 91.13%. Furthermore, the adaptability of DKADE is validated using real-world external clinical data. In summary, DKADE is a powerful tool for studying drug safety and automating adverse event monitoring., (© The Author(s) 2023. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Recent advances and achievements of artificial intelligence (AI) as well as deep and graph learning models have established their usefulness in biomedical applications, especially in drug-drug interactions (DDIs). DDIs refer to a change in the effect of one drug to the presence of another drug in the human body, which plays an essential role in drug discovery and clinical research. DDIs prediction through traditional clinical trials and experiments is an expensive and time-consuming process. To correctly apply the advanced AI and deep learning, the developer and user meet various challenges such as the availability and encoding of data resources, and the design of computational methods. This review summarizes chemical structure based, network based, natural language processing based and hybrid methods, providing an updated and accessible guide to the broad researchers and development community with different domain knowledge. We introduce widely used molecular representation and describe the theoretical frameworks of graph neural network models for representing molecular structures. We present the advantages and disadvantages of deep and graph learning methods by performing comparative experiments. We discuss the potential technical challenges and highlight future directions of deep and graph learning models for accelerating DDIs prediction., (© The Author(s) 2023. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Advancing spatially resolved transcriptomics (ST) technologies help biologists comprehensively understand organ function and tissue microenvironment. Accurate spatial domain identification is the foundation for delineating genome heterogeneity and cellular interaction. Motivated by this perspective, a graph deep learning (GDL) based spatial clustering approach is constructed in this paper. First, the deep graph infomax module embedded with residual gated graph convolutional neural network is leveraged to address the gene expression profiles and spatial positions in ST. Then, the Bayesian Gaussian mixture model is applied to handle the latent embeddings to generate spatial domains. Designed experiments certify that the presented method is superior to other state-of-the-art GDL-enabled techniques on multiple ST datasets. The codes and dataset used in this manuscript are summarized at https://github.com/narutoten520/SCGDL., (© The Author(s) 2023. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

The n -octanol/buffer solution distribution coefficient at pH = 7.4 (log  D 7.4 ) is an indicator of lipophilicity, and it influences a wide variety of absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties and druggability of compounds. In log  D 7.4 prediction, graph neural networks (GNNs) can uncover subtle structure-property relationships (SPRs) by automatically extracting features from molecular graphs that facilitate the learning of SPRs, but their performances are often limited by the small size of available datasets. Herein, we present a transfer learning strategy called pretraining on computational data and then fine-tuning on experimental data (PCFE) to fully exploit the predictive potential of GNNs. PCFE works by pretraining a GNN model on 1.71 million computational log  D data (low-fidelity data) and then fine-tuning it on 19,155 experimental log  D 7.4 data (high-fidelity data). The experiments for three GNN architectures (graph convolutional network (GCN), graph attention network (GAT), and Attentive FP) demonstrated the effectiveness of PCFE in improving GNNs for log  D 7.4 predictions. Moreover, the optimal PCFE-trained GNN model (cx-Attentive FP, R test 2 = 0.909) outperformed four excellent descriptor-based models (random forest (RF), gradient boosting (GB), support vector machine (SVM), and extreme gradient boosting (XGBoost)). The robustness of the cx-Attentive FP model was also confirmed by evaluating the models with different training data sizes and dataset splitting strategies. Therefore, we developed a webserver and defined the applicability domain for this model. The webserver (http://tools.scbdd.com/chemlogd/) provides free log  D 7.4 prediction services. In addition, the important descriptors for log  D 7.4 were detected by the Shapley additive explanations (SHAP) method, and the most relevant substructures of log  D 7.4 were identified by the attention mechanism. Finally, the matched molecular pair analysis (MMPA) was performed to summarize the contributions of common chemical substituents to log  D 7.4 , including a variety of hydrocarbon groups, halogen groups, heteroatoms, and polar groups. In conclusion, we believe that the cx-Attentive FP model can serve as a reliable tool to predict log  D 7.4 and hope that pretraining on low-fidelity data can help GNNs make accurate predictions of other endpoints in drug discovery.

Identification and validation of bioactive small-molecule targets is a significant challenge in drug discovery. In recent years, various in-silico approaches have been proposed to expedite time- and resource-consuming experiments for target detection. Herein, we developed several chemogenomic models for target prediction based on multi-scale information of chemical structures and protein sequences. By combining the information of a compound with multiple protein targets together and putting these compound-target pairs into a well-established model, the scores to indicate whether there are interactions between compounds and targets can be derived, and thus a target prediction task can be completed by sorting the outputted scores. To improve the prediction performance, we constructed several chemogenomic models using multi-scale information of chemical structures and protein sequences, and the ensemble model with the best performance was used as our final model. The model was validated by various strategies and external datasets and the promising target prediction capability of the model, i.e., the fraction of known targets identified in the top-k (1 to 10) list of the potential target candidates suggested by the model, was confirmed. Compared with multiple state-of-art target prediction methods, our model showed equivalent or better predictive ability in terms of the top-k predictions. It is expected that our method can be utilized as a powerful computational tool to narrow down the potential targets for experimental testing., (© 2023. The Author(s).)

Identification of potential targets for known bioactive compounds and novel synthetic analogs is of considerable significance. In silico target fishing (TF) has become an alternative strategy because of the expensive and laborious wet-lab experiments, explosive growth of bioactivity data and rapid development of high-throughput technologies. However, these TF methods are based on different algorithms, molecular representations and training datasets, which may lead to different results when predicting the same query molecules. This can be confusing for practitioners in practical applications. Therefore, this study systematically evaluated nine popular ligand-based TF methods based on target and ligand-target pair statistical strategies, which will help practitioners make choices among multiple TF methods. The evaluation results showed that SwissTargetPrediction was the best method to produce the most reliable predictions while enriching more targets. High-recall similarity ensemble approach (SEA) was able to find real targets for more compounds compared with other TF methods. Therefore, SwissTargetPrediction and SEA can be considered as primary selection methods in future studies. In addition, the results showed that k = 5 was the optimal number of experimental candidate targets. Finally, a novel ensemble TF method based on consensus voting is proposed to improve the prediction performance. The precision of the ensemble TF method outperforms the individual TF method, indicating that the ensemble TF method can more effectively identify real targets within a given top-k threshold. The results of this study can be used as a reference to guide practitioners in selecting the most effective methods in computational drug discovery., (© The Author(s) 2023. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Machine learning-based scoring functions (MLSFs) have become a very favorable alternative to classical scoring functions because of their potential superior screening performance. However, the information of negative data used to construct MLSFs was rarely reported in the literature, and meanwhile the putative inactive molecules recorded in existing databases usually have obvious bias from active molecules. Here we proposed an easy-to-use method named AMLSF that combines active learning using negative molecular selection strategies with MLSF, which can iteratively improve the quality of inactive sets and thus reduce the false positive rate of virtual screening. We chose energy auxiliary terms learning as the MLSF and validated our method on eight targets in the diverse subset of DUD-E. For each target, we screened the IterBioScreen database by AMLSF and compared the screening results with those of the four control models. The results illustrate that the number of active molecules in the top 1000 molecules identified by AMLSF was significantly higher than those identified by the control models. In addition, the free energy calculation results for the top 10 molecules screened out by the AMLSF, null model and control models based on DUD-E also proved that more active molecules can be identified, and the false positive rate can be reduced by AMLSF., (© The Author(s) 2023. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Hematotoxicity has been becoming a serious but overlooked toxicity in drug discovery. However, only a few in silico models have been reported for the prediction of hematotoxicity. In this study, we constructed a high-quality dataset comprising 759 hematotoxic compounds and 1623 nonhematotoxic compounds and then established a series of classification models based on a combination of seven machine learning (ML) algorithms and nine molecular representations. The results based on two data partitioning strategies and applicability domain (AD) analysis illustrate that the best prediction model based on Attentive FP yielded a balanced accuracy (BA) of 72.6%, an area under the receiver operating characteristic curve (AUC) value of 76.8% for the validation set, and a BA of 69.2%, an AUC of 75.9% for the test set. In addition, compared with existing filtering rules and models, our model achieved the highest BA value of 67.5% for the external validation set. Additionally, the shapley additive explanation (SHAP) and atom heatmap approaches were utilized to discover the important features and structural fragments related to hematotoxicity, which could offer helpful tips to detect undesired positive substances. Furthermore, matched molecular pair analysis (MMPA) and representative substructure derivation technique were employed to further characterize and investigate the transformation principles and distinctive structural features of hematotoxic chemicals. We believe that the novel graph-based deep learning algorithms and insightful interpretation presented in this study can be used as a trustworthy and effective tool to assess hematotoxicity in the development of new drugs.

Traditional Chinese Medicine (TCM) has been widely used in the treatment of various diseases for millennia. In the modernization process of TCM, TCM ingredient databases are playing more and more important roles. However, most of the existing TCM ingredient databases do not provide simplification function for extracting key ingredients in each herb or formula, which hinders the research on the mechanism of actions of the ingredients in TCM databases. The lack of quality control and standardization of the data in most of these existing databases is also a prominent disadvantage. Therefore, we developed a Traditional Chinese Medicine Simplified Integrated Database (TCMSID) with high storage, high quality and standardization. The database includes 499 herbs registered in the Chinese pharmacopeia with 20,015 ingredients, 3270 targets as well as corresponding detailed information. TCMSID is not only a database of herbal ingredients, but also a TCM simplification platform. Key ingredients from TCM herbs are available to be screened out and regarded as representatives to explore the mechanism of TCM herbs by implementing multi-tool target prediction and multilevel network construction. TCMSID provides abundant data sources and analysis platforms for TCM simplification and drug discovery, which is expected to promote modernization and internationalization of TCM and enhance its international status in the future. TCMSID is freely available at https://tcm.scbdd.com ., (© 2022. The Author(s).)

Accurate prediction of pharmacological properties of small molecules is becoming increasingly important in drug discovery. Traditional feature-engineering approaches heavily rely on handcrafted descriptors and/or fingerprints, which need extensive human expert knowledge. With the rapid progress of artificial intelligence technology, data-driven deep learning methods have shown unparalleled advantages over feature-engineering-based methods. However, existing deep learning methods usually suffer from the scarcity of labeled data and the inability to share information between different tasks when applied to predicting molecular properties, thus resulting in poor generalization capability. Here, we proposed a novel multitask learning BERT (Bidirectional Encoder Representations from Transformer) framework, named MTL-BERT, which leverages large-scale pre-training, multitask learning, and SMILES (simplified molecular input line entry specification) enumeration to alleviate the data scarcity problem. MTL-BERT first exploits a large amount of unlabeled data through self-supervised pretraining to mine the rich contextual information in SMILES strings and then fine-tunes the pretrained model for multiple downstream tasks simultaneously by leveraging their shared information. Meanwhile, SMILES enumeration is used as a data enhancement strategy during the pretraining, fine-tuning, and test phases to substantially increase data diversity and help to learn the key relevant patterns from complex SMILES strings. The experimental results showed that the pretrained MTL-BERT model with few additional fine-tuning can achieve much better performance than the state-of-the-art methods on most of the 60 practical molecular datasets. Additionally, the MTL-BERT model leverages attention mechanisms to focus on SMILES character features essential to target properties for model interpretability., (Copyright © 2022 Xiao-Chen Zhang et al.)

Decaprenylphosphoryl-β-D-ribose oxidase (DprE1) plays important roles in the biosynthesis of mycobacterium cell wall. DprE1 inhibitors have shown great potentials in the development of new regimens for tuberculosis (TB) treatment. In this study, an integrated molecular modeling strategy, which combined computational bioactivity fingerprints and structure-based virtual screening, was employed to identify potential DprE1 inhibitors. Two lead compounds (B2 and H3) that could inhibit DprE1 and thus kill Mycobacterium smegmatis in vitro were identified. Moreover, compound H3 showed potent inhibitory activity against Mycobacterium tuberculosis in vitro (MIC Mtb  = 1.25 μM) and low cytotoxicity against mouse embryo fibroblast NIH-3T3 cells. Our research provided an effective strategy to discover novel anti-TB lead compounds., (© 2021. The Author(s), under exclusive licence to CPS and SIMM.)

Physapubenolide (PB), a withanolide-type compound extracted from the traditional herb Physalis minima L., has been demonstrated to exert remarkable cytotoxicity against cancer cells; however, its molecular mechanisms are still unclear. In this study, we demonstrated that PB inhibited cell proliferation and migration in melanoma cells by inducing cell apoptosis. The anticancer activity of PB was further verified in a melanoma xenograft model. To explore the mechanism underlying the anticancer effects of PB, we carried out an in silico target prediction study, which combined three approaches (chemical similarity searching, quantitative structure-activity relationship (QSAR), and molecular docking) to identify the targets of PB, and found that PB likely targets 3-hydroxy-methylglutaryl CoA reductase (HMGCR), the rate-limiting enzyme of the mevalonate pathway, which promotes cancer cell proliferation, migration, and metastasis. We further demonstrated that PB interacted with HMGCR, decreased its protein expression and inhibited the HMGCR/YAP pathway in melanoma cells. In addition, we found that PB could restore vemurafenib sensitivity in vemurafenib-resistant A-375 cells, which was correlated with the downregulation of HMGCR. In conclusion, we demonstrate that PB elicits anticancer action and enhances sensitivity to vemurafenib by targeting HMGCR., (© 2021. The Author(s).)

Drug-drug interaction (DDI) often causes serious adverse reactions and thus results in inestimable economic and social loss. Currently, comprehensive DDI evaluation has become a major challenge in pharmaceutical research due to the time-consuming and costly process of the experimental assessment and it is of high necessity to develop effective in silico methods to predict and evaluate DDIs accurately and efficiently. In this study, based on a large number of substrates and inhibitors related to five important CYP450 isozymes (CYP1A2, CYP2C9, CYP2C19, CYP2D6 and CYP3A4), a series of high-performance predictive models for metabolic DDIs were constructed by two machine learning methods (random forest and XGBoost) and 4 different types of descriptors (MOE&#95;2D, CATS, ECFP4 and MACCS). To reduce the uncertainty of individual models, the consensus method was applied to yield more reliable predictions. A series of evaluations illustrated that the consensus models were more reliable and robust for the DDI predictions of new drug combination. For the internal validation, the whole prediction accuracy and AUC value of the DDI models were around 0.8 and 0.9, respectively. When it was applied to the external datasets, the model accuracy was 0.793 and 0.795 for multi-level validation and external validation, respectively. Furthermore, we also compared our model with some recently published tools and then applied the final model to predict FDA-approved drugs and proposed 54,013 possible drug pairs with potential DDIs. In summary, we developed a powerful DDI predictive model from the perspective of the CYP450 enzyme family and it will help a lot in the future drug development and clinical pharmacy research., (© 2022. The Author(s).)

Nowadays, computational approaches have drawn more and more attention when exploring the relationship between sweetness and chemical structure instead of traditional experimental tests. In this work, we proposed a novel multi-layer sweetness evaluation system based on machine learning methods. It can be used to evaluate sweet properties of compounds with different chemical spaces and categories, including natural, artificial, carbohydrate, non-carbohydrate, nutritive and non-nutritive ones, suitable for different application scenarios. Furthermore, it provided quantitative predictions of sweetness. In addition, sweetness-related chemical basis and structure transforming rules were obtained by using molecular cloud and matched molecular pair analysis (MMPA) methods. This work systematically improved the data quality, explored the best machine learning algorithm and molecular characterizing strategy, and finally obtained robust models to establish a multi-layer prediction system (available at: https://github.com/ifyoungnet/ChemSweet). We hope that this study could facilitate food scientists with efficient screening and precise development of high-quality sweeteners., (Copyright © 2021 Elsevier Ltd. All rights reserved.)

Structural information for chemical compounds is often described by pictorial images in most scientific documents, which cannot be easily understood and manipulated by computers. This dilemma makes optical chemical structure recognition (OCSR) an essential tool for automatically mining knowledge from an enormous amount of literature. However, existing OCSR methods fall far short of our expectations for realistic requirements due to their poor recovery accuracy. In this paper, we developed a deep neural network model named ABC-Net (Atom and Bond Center Network) to predict graph structures directly. Based on the divide-and-conquer principle, we propose to model an atom or a bond as a single point in the center. In this way, we can leverage a fully convolutional neural network (CNN) to generate a series of heat-maps to identify these points and predict relevant properties, such as atom types, atom charges, bond types and other properties. Thus, the molecular structure can be recovered by assembling the detected atoms and bonds. Our approach integrates all the detection and property prediction tasks into a single fully CNN, which is scalable and capable of processing molecular images quite efficiently. Experimental results demonstrate that our method could achieve a significant improvement in recognition performance compared with publicly available tools. The proposed method could be considered as a promising solution to OCSR problems and a starting point for the acquisition of molecular information in the literature., (© The Author(s) 2022. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Motivation: Understanding chemical-gene interactions (CGIs) is crucial for screening drugs. Wet experiments are usually costly and laborious, which limits relevant studies to a small scale. On the contrary, computational studies enable efficient in-silico exploration. For the CGI prediction problem, a common method is to perform systematic analyses on a heterogeneous network involving various biomedical entities. Recently, graph neural networks become popular in the field of relation prediction. However, the inherent heterogeneous complexity of biological interaction networks and the massive amount of data pose enormous challenges. This paper aims to develop a data-driven model that is capable of learning latent information from the interaction network and making correct predictions., Results: We developed BioNet, a deep biological networkmodel with a graph encoder-decoder architecture. The graph encoder utilizes graph convolution to learn latent information embedded in complex interactions among chemicals, genes, diseases and biological pathways. The learning process is featured by two consecutive steps. Then, embedded information learnt by the encoder is then employed to make multi-type interaction predictions between chemicals and genes with a tensor decomposition decoder based on the RESCAL algorithm. BioNet includes 79 325 entities as nodes, and 34 005 501 relations as edges. To train such a massive deep graph model, BioNet introduces a parallel training algorithm utilizing multiple Graphics Processing Unit (GPUs). The evaluation experiments indicated that BioNet exhibits outstanding prediction performance with a best area under Receiver Operating Characteristic (ROC) curve of 0.952, which significantly surpasses state-of-theart methods. For further validation, top predicted CGIs of cancer and COVID-19 by BioNet were verified by external curated data and published literature., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

In the process of drug discovery, the optimization of lead compounds has always been a challenge faced by pharmaceutical chemists. Matched molecular pair analysis (MMPA), a promising tool to efficiently extract and summarize the relationship between structural transformation and property change, is suitable for local structural optimization tasks. Especially, the integration of MMPA with QSAR modeling can further strengthen the utility of MMPA in molecular optimization navigation. In this study, a new semi-automated procedure based on KNIME was developed to support MMPA on both large- and small-scale datasets, including molecular preparation, QSAR model construction, applicability domain evaluation, and MMP calculation and application. Two examples covering regression and classification tasks were provided to gain a better understanding of the importance of MMPA, which has also shown the reliability and utility of this MMPA-by-QSAR pipeline., (© 2021. The Author(s).)

Computational methods have become indispensable tools to accelerate the drug discovery process and alleviate the excessive dependence on time-consuming and labor-intensive experiments. Traditional feature-engineering approaches heavily rely on expert knowledge to devise useful features, which could be costly and sometimes biased. The emerging deep learning (DL) methods deliver a data-driven method to automatically learn expressive representations from complex raw data. Inspired by this, researchers have attempted to apply various deep neural network models to simplified molecular input line entry specification (SMILES) strings, which contain all the composition and structure information of molecules. However, current models usually suffer from the scarcity of labeled data. This results in a low generalization ability of SMILES-based DL models, which prevents them from competing with the state-of-the-art computational methods. In this study, we utilized the BiLSTM (bidirectional long short term merory) attention network (BAN) in which we employed a novel multi-step attention mechanism to facilitate the extracting of key features from the SMILES strings. Meanwhile, SMILES enumeration was utilized as a data augmentation method in the training phase to substantially increase the number of labeled data and enlarge the probability of mining more patterns from complex SMILES. We again took advantage of SMILES enumeration in the prediction phase to rectify model prediction bias and provide a more accurate prediction. Combined with the BAN model, our strategies can greatly improve the performance of latent features learned from SMILES strings. In 11 canonical absorption, distribution, metabolism, excretion and toxicity-related tasks, our method outperformed the state-of-the-art approaches., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

The poly (ADP-ribose) polymerase-1 (PARP1) has been regarded as a vital target in recent years and PARP1 inhibitors can be used for ovarian and breast cancer therapies. However, it has been realized that most of PARP1 inhibitors have disadvantages of low solubility and permeability. Therefore, by discovering more molecules with novel frameworks, it would have greater opportunities to apply it into broader clinical fields and have a more profound significance. In the present study, multiple virtual screening (VS) methods had been employed to evaluate the screening efficiency of ligand-based, structure-based and data fusion methods on PARP1 target. The VS methods include 2D similarity screening, structure-activity relationship (SAR) models, docking and complex-based pharmacophore screening. Moreover, the sum rank, sum score and reciprocal rank were also adopted for data fusion methods. The evaluation results show that the similarity searching based on Torsion fingerprint, six SAR models, Glide docking and pharmacophore screening using Phase have excellent screening performance. The best data fusion method is the reciprocal rank, but the sum score also performs well in framework enrichment. In general, the ligand-based VS methods show better performance on PARP1 inhibitor screening. These findings confirmed that adding ligand-based methods to the early screening stage will greatly improve the screening efficiency, and be able to enrich more highly active PARP1 inhibitors with diverse structures., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Motivation: Accurate and efficient prediction of molecular properties is one of the fundamental issues in drug design and discovery pipelines. Traditional feature engineering-based approaches require extensive expertise in the feature design and selection process. With the development of artificial intelligence (AI) technologies, data-driven methods exhibit unparalleled advantages over the feature engineering-based methods in various domains. Nevertheless, when applied to molecular property prediction, AI models usually suffer from the scarcity of labeled data and show poor generalization ability., Results: In this study, we proposed molecular graph BERT (MG-BERT), which integrates the local message passing mechanism of graph neural networks (GNNs) into the powerful BERT model to facilitate learning from molecular graphs. Furthermore, an effective self-supervised learning strategy named masked atoms prediction was proposed to pretrain the MG-BERT model on a large amount of unlabeled data to mine context information in molecules. We found the MG-BERT model can generate context-sensitive atomic representations after pretraining and transfer the learned knowledge to the prediction of a variety of molecular properties. The experimental results show that the pretrained MG-BERT model with a little extra fine-tuning can consistently outperform the state-of-the-art methods on all 11 ADMET datasets. Moreover, the MG-BERT model leverages attention mechanisms to focus on atomic features essential to the target property, providing excellent interpretability for the trained model. The MG-BERT model does not require any hand-crafted feature as input and is more reliable due to its excellent interpretability, providing a novel framework to develop state-of-the-art models for a wide range of drug discovery tasks., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Matched molecular pairs analysis (MMPA) has become a powerful tool for automatically and systematically identifying medicinal chemistry transformations from compound/property datasets. However, accurate determination of matched molecular pair (MMP) transformations largely depend on the size and quality of existing experimental data. Lack of high-quality experimental data heavily hampers the extraction of more effective medicinal chemistry knowledge. Here, we developed a new strategy called quantitative structure-activity relationship (QSAR)-assisted-MMPA to expand the number of chemical transformations and took the logD7.4 property endpoint as an example to demonstrate the reliability of the new method. A reliable logD7.4 consensus prediction model was firstly established, and its applicability domain was strictly assessed. By applying the reliable logD7.4 prediction model to screen two chemical databases, we obtained more high-quality logD7.4 data by defining a strict applicability domain threshold. Then, MMPA was performed on the predicted data and experimental data to derive more chemical rules. To validate the reliability of the chemical rules, we compared the magnitude and directionality of the property changes of the predicted rules with those of the measured rules. Then, we compared the novel chemical rules generated by our proposed approach with the published chemical rules, and found that the magnitude and directionality of the property changes were consistent, indicating that the proposed QSAR-assisted-MMPA approach has the potential to enrich the collection of rule types or even identify completely novel rules. Finally, we found that the number of the MMP rules derived from the experimental data could be amplified by the predicted data, which is helpful for us to analyze the medicinal chemical rules in local chemical environment. In summary, the proposed QSAR-assisted-MMPA approach could be regarded as a very promising strategy to expand the chemical transformation space for lead optimization, especially when no enough experimental data can support MMPA., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Background: Substructure screening is widely applied to evaluate the molecular potency and ADMET properties of compounds in drug discovery pipelines, and it can also be used to interpret QSAR models for the design of new compounds with desirable physicochemical and biological properties. With the continuous accumulation of more experimental data, data-driven computational systems which can derive representative substructures from large chemical libraries attract more attention. Therefore, the development of an integrated and convenient tool to generate and implement representative substructures is urgently needed., Results: In this study, PySmash, a user-friendly and powerful tool to generate different types of representative substructures, was developed. The current version of PySmash provides both a Python package and an individual executable program, which achieves ease of operation and pipeline integration. Three types of substructure generation algorithms, including circular, path-based and functional group-based algorithms, are provided. Users can conveniently customize their own requirements for substructure size, accuracy and coverage, statistical significance and parallel computation during execution. Besides, PySmash provides the function for external data screening., Conclusion: PySmash, a user-friendly and integrated tool for the automatic generation and implementation of representative substructures, is presented. Three screening examples, including toxicophore derivation, privileged motif detection and the integration of substructures with machine learning (ML) models, are provided to illustrate the utility of PySmash in safety profile evaluation, therapeutic activity exploration and molecular optimization, respectively. Its executable program and Python package are available at https://github.com/kotori-y/pySmash., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Background: Fluorescent detection methods are indispensable tools for chemical biology. However, the frequent appearance of potential fluorescent compound has greatly interfered with the recognition of compounds with genuine activity. Such fluorescence interference is especially difficult to identify as it is reproducible and possesses concentration-dependent characteristic. Therefore, the development of a credible screening tool to detect fluorescent compounds from chemical libraries is urgently needed in early stages of drug discovery., Results: In this study, we developed a webserver ChemFLuo for fluorescent compound detection, based on two large and high-quality training datasets containing 4906 blue and 8632 green fluorescent compounds. These molecules were used to construct a group of prediction models based on the combination of three machine learning algorithms and seven types of molecular representations. The best blue fluorescence prediction model achieved with balanced accuracy (BA) = 0.858 and area under the receiver operating characteristic curve (AUC) = 0.931 for the validation set, and BA = 0.823 and AUC = 0.903 for the test set. The best green fluorescence prediction model achieved the prediction accuracy with BA = 0.810 and AUC = 0.887 for the validation set, and BA = 0.771 and AUC = 0.852 for the test set. Besides prediction model, 22 blue and 16 green representative fluorescent substructures were summarized for the screening of potential fluorescent compounds. The comparison with other fluorescence detection tools and theapplication to external validation sets and large molecule libraries have demonstrated the reliability of prediction model for fluorescent compound detection., Conclusion: ChemFLuo is a public webserver to filter out compounds with undesirable fluorescent properties, which will benefit the design of high-quality chemical libraries for drug discovery. It is freely available at http://admet.scbdd.com/chemfluo/index/., (© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

As one of the central tasks of modern medicinal chemistry, scaffold hopping is expected to lead to the discovery of structural novel biological active compounds and broaden the chemical space of known active compounds. Here, we report the computational bioactivity fingerprint (CBFP) for easier scaffold hopping, where the predicted activities in multiple quantitative structure-activity relationship models are integrated to characterize the biological space of a molecule. In retrospective benchmarks, the CBFP representation shows outstanding scaffold hopping potential relative to other chemical descriptors. In the prospective validation for the discovery of novel inhibitors of poly [ADP-ribose] polymerase 1, 35 predicted compounds with diverse structures are tested, 25 of which show detectable growth-inhibitory activity; beyond this, the most potent (compound 6) has an IC 50 of 0.263 nM. These results support the use of CBFP representation as the bioactivity proxy of molecules to explore uncharted chemical space and discover novel compounds.

In 2010, the pan-assay interference compounds (PAINS) rule was proposed to identify false-positive compounds, especially frequent hitters (FHs), in biological screening campaigns, and has rapidly become an essential component in drug design. However, the specific mechanisms remain unknown, and the result validation and follow-up processing schemes are still unclear. In this review, a large benchmark collection of >600,000 compounds sourced from databases and the literature, including six common false-positive mechanisms, was used to evaluate the detection ability of PAINS. In addition, 400 million purchasable molecules from the ZINC database were also applied to PAINS screening. The results indicate that the PAINS rule is not suitable for the screening of all types of false-positive results and needs more improvement., (Copyright © 2021 Elsevier Ltd. All rights reserved.)

Scoring functions (SFs) based on complex machine learning (ML) algorithms have gradually emerged as a promising alternative to overcome the weaknesses of classical SFs. However, extensive efforts have been devoted to the development of SFs based on new protein-ligand interaction representations and advanced alternative ML algorithms instead of the energy components obtained by the decomposition of existing SFs. Here, we propose a new method named energy auxiliary terms learning (EATL), in which the scoring components are extracted and used as the input for the development of three levels of ML SFs including EATL SFs, docking-EATL SFs and comprehensive SFs with ascending VS performance. The EATL approach not only outperforms classical SFs for the absolute performance (ROC) and initial enrichment (BEDROC) but also yields comparable performance compared with other advanced ML-based methods on the diverse subset of Directory of Useful Decoys: Enhanced (DUD-E). The test on the relatively unbiased actives as decoys (AD) dataset also proved the effectiveness of EATL. Furthermore, the idea of learning from SF components to yield improved screening power can also be extended to other docking programs and SFs available., (© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Background: High-throughput screening (HTS) and virtual screening (VS) have been widely used to identify potential hits from large chemical libraries. However, the frequent occurrence of 'noisy compounds' in the screened libraries, such as compounds with poor drug-likeness, poor selectivity or potential toxicity, has greatly weakened the enrichment capability of HTS and VS campaigns. Therefore, the development of comprehensive and credible tools to detect noisy compounds from chemical libraries is urgently needed in early stages of drug discovery., Results: In this study, we developed a freely available integrated python library for negative design, called Scopy, which supports the functions of data preparation, calculation of descriptors, scaffolds and screening filters, and data visualization. The current version of Scopy can calculate 39 basic molecular properties, 3 comprehensive molecular evaluation scores, 2 types of molecular scaffolds, 6 types of substructure descriptors and 2 types of fingerprints. A number of important screening rules are also provided by Scopy, including 15 drug-likeness rules (13 drug-likeness rules and 2 building block rules), 8 frequent hitter rules (four assay interference substructure filters and four promiscuous compound substructure filters), and 11 toxicophore filters (five human-related toxicity substructure filters, three environment-related toxicity substructure filters and three comprehensive toxicity substructure filters). Moreover, this library supports four different visualization functions to help users to gain a better understanding of the screened data, including basic feature radar chart, feature-feature-related scatter diagram, functional group marker gram and cloud gram., Conclusion: Scopy provides a comprehensive Python package to filter out compounds with undesirable properties or substructures, which will benefit the design of high-quality chemical libraries for drug design and discovery. It is freely available at https://github.com/kotori-y/Scopy., (© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Background: With the increasing development of biotechnology and information technology, publicly available data in chemistry and biology are undergoing explosive growth. Such wealthy information in these resources needs to be extracted and then transformed to useful knowledge by various data mining methods. However, a main computational challenge is how to effectively represent or encode molecular objects under investigation such as chemicals, proteins, DNAs and even complicated interactions when data mining methods are employed. To further explore these complicated data, an integrated toolkit to represent different types of molecular objects and support various data mining algorithms is urgently needed., Results: We developed a freely available R/CRAN package, called BioMedR, for molecular representations of chemicals, proteins, DNAs and pairwise samples of their interactions. The current version of BioMedR could calculate 293 molecular descriptors and 13 kinds of molecular fingerprints for small molecules, 9920 protein descriptors based on protein sequences and six types of generalized scale-based descriptors for proteochemometric modeling, more than 6000 DNA descriptors from nucleotide sequences and six types of interaction descriptors using three different combining strategies. Moreover, this package realized five similarity calculation methods and four powerful clustering algorithms as well as several useful auxiliary tools, which aims at building an integrated analysis pipeline for data acquisition, data checking, descriptor calculation and data modeling., Conclusion: BioMedR provides a comprehensive and uniform R package to link up different representations of molecular objects with each other and will benefit cheminformatics/bioinformatics and other biomedical users. It is available at: https://CRAN.R-project.org/package=BioMedR and https://github.com/wind22zhu/BioMedR/., (© The Author(s) 2019. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Organophosphates (OPs) are highly toxic compounds, with widespread application in agricultural and chemical industries, whose introduction into the environment poses serious hazards to humans and ecological systems. To assess and ultimately mitigate these hazards, this study predicted the acute toxicity of OPs according to their chemical structure and administration route. The acute toxicity data of 161 OPs in two species via six different administration routes were manually collected and used to develop a series of quantitative structure-toxicity relationship (QSTR) models with robust and practical predictive abilities. The random forest algorithm was used to develop the models, employing both quantum chemical and two-dimensional descriptors according to OECD guidelines. Correlation results and feature similarities indicated that whereas acute toxicity data from rats and mice via the same administration route were combinable for modeling, data from different routes were not. Six QSTR models for each route in a single species and two QSTR models for a single route in the two species were constructed, achieving practical predictive performance. Despite significant variances in their datasets, the prediction models could predict the acute toxicity of novel or unknown OPs, realize rapid assessment, and provide guidance for regulatory decisions to reduce the hazards of OPs., (Copyright © 2020 Elsevier B.V. All rights reserved.)

Natural products, as an ideal starting point for molecular design, play a pivotal role in drug discovery; however, ambiguous targets and mechanisms have limited their in-depth research and applications in a global dimension. In-silico target prediction methods have become an alternative to target identification experiments due to the high accuracy and speed, but most studies only use a single prediction method, which may reduce the accuracy and reliability of the prediction. Here, we firstly presented a combinatorial target screening strategy to facilitate multi-target screening of natural products considering the characteristics of diverse in-silico target prediction methods, which consists of ligand-based online approaches, consensus SAR modelling and target-specific re-scoring function modelling. To validate the practicability of the strategy, natural product neferine, a bisbenzylisoquinoline alkaloid isolated from the lotus seed, was taken as an example to illustrate the screening process and a series of corresponding experiments were implemented to explore the pharmacological mechanisms of neferine. The proposed computational method could be used for a complementary hypothesis generation and rapid analysis of potential targets of natural products., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier Masson SAS. All rights reserved.)

Autophagy can protect stressed cancer cell by degradation of damaged proteins and organelles. However, the regulatory mechanisms behind this cellular process remain incompletely understood. Here, we demonstrate that RSK2 (p90 ribosomal S6 kinase 2) plays a critical role in ER stress-induced autophagy in breast cancer cells. We demonstrated that the promotive effect of RSK2 on autophagy resulted from directly binding of AMPKα2 in nucleus and phosphorylating it at Thr172 residue. IRE1α, an ER membrane-associated protein mediating unfolded protein response (UPR), is required for transducing the signal for activation of ERK1/2-RSK2 under ER stress. Suppression of autophagy by knockdown of RSK2 enhanced the sensitivity of breast cancer cells to ER stress both in vitro and in vivo. Furthermore, we demonstrated that inhibition of RSK2-mediated autophagy rendered breast cancer cells more sensitive to paclitaxel, a chemotherapeutic agent that induces ER stress-mediated cell death. This study identifies RSK2 as a novel controller of autophagy in tumor cells and suggests that targeting RSK2 can be exploited as an approach to reinforce the efficacy of ER stress-inducing agents against cancer.

Poly(ADP-ribose) Polymerase 1 (PARP1), one of the most investigated 18 membered PARP family enzymes, is involved in a variety of cellular functions including DNA damage repair, gene transcription and cell apoptosis. PARP1 can form a PARP1(ADP-ribose) polymers, then bind to the DNA damage gap to recruit DNA repair proteins, and repair the break to maintain genomic stability. PARP1 is highly expressed in tumor cells, so the inhibition of PARP1 can block DNA repair, promote tumor cell apoptosis, and exert antitumor activity. To date, four PARP1 inhibitors namely olaparib, rucaparib, niraparib and talazoparib, have been approved by Food and Drug Administration (FDA) for treating ovarian cancer and breast cancer with BRCA1/2 mutation. These drugs have showed super advantages over conventional chemotherapeutic drugs with low hematological toxicity and slowly developed drug resistance. In this article, we summarize and analyze the structure features of PARP1, the biological functions and antitumor mechanisms of PARP1 inhibitors. Importantly, we suggest that establishing a new structure-activity relationship of developed PARP1 inhibitors via substructural searching and the matched molecular pair analysis would accelerate the process in finding more potent and safer PARP1 inhibitors., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier Masson SAS. All rights reserved.)

Virtual Screening (VS) based on molecular docking is an efficient method used for retrieving novel hit compounds in drug discovery. However, the accuracy of the current docking scoring function (SF) is usually insufficient. In this study, in order to improve the screening power of SF, a novel approach named EAT-Score was proposed by directly utilizing the energy auxiliary terms (EAT) provided by molecular docking scoring through eXtreme Gradient Boosting (XGBoost). Here, EAT specifically refers to the output of the Molecular Operating Environment (MOE) scoring, including the energy scores of five different classical SFs and the Protein-Ligand Interaction Fingerprint (PLIF) terms. The performance of EAT-Score to discriminate actives from decoys was strictly validated on the DUD-E diverse subset by using different performance metrics. The results showed that EAT-Score performed much better than classical SFs in VS, with its AUC values exhibiting an improvement of around 0.3. Meanwhile, EAT-Score could achieve comparable even better prediction performance compared with other state-of-the-art VS methods, such as some machine learning (ML)-based SFs and classical SFs implemented in docking programs, in terms of AUC, LogAUC, or BEDROC. Furthermore, the EAT-Score model can capture important binding pattern information from protein-ligand complexes by Shapley additive explanations (SHAP) analysis, which may be very helpful in interpreting the ligand binding mechanism for a certain target and thereby guiding drug design.

It has been realized that FDA approved drugs may have more molecular targets than is commonly thought. Thus, to find the exact drug-target interactions (DTIs) is of great significance for exploring the new molecular mechanism of drugs. Here, we developed a multi-scale system pharmacology (MSSP) method for the large-scale prediction of DTIs. We used MSSP to integrate drug-related and target-related data from multiple levels, the network structural data formed by known drug-target relationships for predicting likely unknown DTIs. Prediction results revealed that Ixabepilone, an epothilone B analog for treating breast cancer patients, may target Bcl-2, an oncogene that contributes to tumor progression and therapy resistance by inhibiting apoptosis. Furthermore, we demonstrated that Ixabepilone could bind with Bcl-2 and decrease its protein expression in breast cancer cells. The down-regulation of Bcl-2 by Ixabepilone is resulted from promoting its degradation by affecting p-Bcl-2. We further found that Ixabepilone could induce autophagy by releasing Beclin1 from Beclin1/Bcl-2 complex. Inhibition of autophagy by knockdown of Beclin1 or pharmacological inhibitor augmented apoptosis, thus enhancing the antitumor efficacy of Ixabepilone against breast cancer cells in vitro and in vivo. In addition, Ixabepilone also decreases Bcl-2 protein expression and induces cytoprotective autophagy in human hepatic carcinoma and glioma cells. In conclusion, this study not only provides a feasible and alternative way exploring new molecular mechanisms of drugs by combing computation DTI prediction, but also reveals an effective strategy to reinforce the antitumor efficacy of Ixabepilone., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier Masson SAS. All rights reserved.)

Aim: To describe the factors affecting distant metastasis of Merkel cell carcinoma (MCC) and the prognosis of metastatic MCC. Materials & methods: The MCC patient information was downloaded from the SEER database. Logistic regression and Cox proportional hazard models were conducted to screen for significant factors. Results: A total of 3449 patients were enrolled. Surgery and chemotherapy were significantly correlated with the occurrence of distant metastasis. In the cause-specific survival rate of MCC, regional lymph node removal, sentinel lymph node biopsy, radiation and chemotherapy can significantly reduce the prognostic risk of patients with distant metastases. Conclusion: Our study screened out the factors affecting the distant metastasis and prognosis of MCC and more prospective studies are needed to verify our findings.

Tyrosinase (TYR) is a crucial enzyme involved in melanogenesis, and its overexpression is closely associated with melanoma. To precisely monitor intracellular TYR activity, remote control of a molecule imaging tool is highly meaningful but remains to be explored. In this work, we present the first photocaged tyrosinase fluorogenic probe by caging the substrate of the enzymatic probe with a photolabile group. Because of the sequential light and enzyme-activation feature, this probe exhibits photocontrollable "turn on" response toward TYR with good selectivity and high sensitivity (detection limit: 0.08 U/mL). Fluorescence imaging results validate that the caged probe possesses the capability of visualizing intracellular endogenous tyrosinase activity in a photocontrol fashion, thus offering a promising molecule imaging tool for investigating TYR-related physiological function and pathological role. Moreover, our sequential activation strategy has great potential for developing more photocontrollable enzymatic fluorogenic probes with spatiotemporal resolution.

Negative design is a group of virtual screening methods that aims at weeding out compounds with undesired properties during the early stages of drug development. These methods are mainly designed to predict three important types of pharmacological properties: drug-likeness, frequent hitters, and toxicity. In order to achieve high screening efficiency, most negative design methods are physicochemical property-based and/or substructure-based rules or filters. Such methods have advantages of simplicity and good interpretability, but they also suffer from some defects such as inflexibility, discontinuity, and hard decision-making. In this review, the advances in negative design for the evaluations of drug-likeness, frequent hitters, and toxicity are outlined. In addition, the related Web servers and software packages developed recently for negative design are summarized. Finally, future research directions in this field are discussed.

Luciferase-based bioluminescence detection techniques are highly favored in high-throughput screening (HTS), in which the firefly luciferase (FLuc) is the most commonly used variant. However, FLuc inhibitors can interfere with the activity of luciferase, which may result in false positive signals in HTS assays. In order to reduce the unnecessary cost of time and money, an in silico prediction model for FLuc inhibitors is highly desirable. In this study, we built an extensive data set consisting of 20 888 FLuc inhibitors and 198 608 noninhibitors, and then developed a group of classification models based on the combination of three machine learning (ML) algorithms and four types of molecular representations. The best prediction model based on XGBoost and ECFP4 and MOE2d descriptors yielded a balanced accuracy (BA) of 0.878 and an area under the receiver operating characteristic curve (AUC) value of 0.958 for the validation set, and a BA of 0.886 and an AUC of 0.947 for the test set. Three external validation sets, including set 1 (3231 FLuc inhibitors and 69 783 noninhibitors), set 2 (695 FLuc inhibitors and 75 913 noninhibitors), and set 3 (1138 FLuc inhibitors and 8155 noninhibitors), were used to verify the predictive ability of our models. The BA values for the three external validation sets given by the best model are 0.864, 0.845, and 0.791, respectively. In addition, the important features or structural fragments related to FLuc inhibitors were recognized by the Shapley additive explanations (SHAP) method along with their influences on predictions, which may provide valuable clues to detecting undesirable luciferase inhibitors. Based on the important and explanatory features, 16 rules were proposed for detecting FLuc inhibitors, which can achieve a correction rate of 70% for FLuc inhibitors. Furthermore, a comparison with existing prediction rules and models for FLuc inhibitors used in virtual screening verified the high reliability of the models and rules proposed in this study. We also used the model to screen three curated chemical databases, and almost 10% of the molecules in the evaluated databases were predicted as inhibitors, highlighting the potential risk of false positives in luciferase-based assays. Finally, a public web server called ChemFLuc was developed (http://admet.scbdd.com/chemfluc/index/), and it offers a free available service to predict potential FLuc inhibitors.

One of the major challenges in early drug discovery is the recognition of frequent hitters (FHs), that is, compounds that nonspecifically bind to a range of macromolecular targets or false positives caused by various types of assay interferences. In this review, we survey the mechanisms underlying different types of FHs, including aggregators, spectroscopic interference compounds (i.e., luciferase inhibitors and fluorescent compounds), chemical reactive compounds, and promiscuous compounds. We also review commonly used experimental detection techniques and computational prediction models for FH identification. In addition, the rational applications of these computational filters are discussed. It is believed that, with the rational use of FH filters, the efficiency of drug discovery will be significantly improved., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Objective: To investigate the potential effects of microRNA-429-5p (miR-429-5p) on the development of malignant melanoma (MM) and the relevant mechanism., Patients and Methods: Quantitative Real Time-Polymerase Chain Reaction (qRT-PCR) was used to detect the differential expression of miR-429-5p in MM tissues. The relationship between miR-429-5p expression and clinical pathological data of MM patients was analyzed. LIM kinase 1 (LIMK1) was verified as a downstream target of miR-429-5p by online prediction software, and the interaction between LIMK1 and miR-429-5p was verified by Dual-Luciferase reporter assay., Results: Compared with normal skin tissues, miR-429-5p was downregulated in MM tissues. MiR-429-5p expression was correlated with tumor size and stage of MM. Upregulation of miR-429-5p significantly inhibited protein expression of LIMK1 and reduced migration and invasion ability of MM cells. LIMK1 was involved in MM progression regulated by miR-429-5p., Conclusions: MiR-429-5p attenuates migration and invasion in MM by targeting LIMK1. Hence, miR-429-5p/LIMK1 axis might be a potential therapeutic target for the treatment of MM.

Lipophilicity, as evaluated by the n -octanol/buffer solution distribution coefficient at pH = 7.4 (log  D 7.4 ), is a major determinant of various absorption, distribution, metabolism, elimination, and toxicology (ADMET) parameters of drug candidates. In this study, we developed several quantitative structure-property relationship (QSPR) models to predict log  D 7.4 based on a large and structurally diverse data set. Eight popular machine learning algorithms were employed to build the prediction models with 43 molecular descriptors selected by a wrapper feature selection method. The results demonstrated that XGBoost yielded better prediction performance than any other single model ( R T 2 = 0.906 and RMSE T = 0.395). Moreover, the consensus model from the top three models could continue to improve the prediction performance ( R T 2 = 0.922 and RMSE T = 0.359). The robustness, reliability, and generalization ability of the models were strictly evaluated by the Y-randomization test and applicability domain analysis. Moreover, the group contribution model based on 110 atom types and the local models for different ionization states were also established and compared to the global models. The results demonstrated that the descriptor-based consensus model is superior to the group contribution method, and the local models have no advantage over the global models. Finally, matched molecular pair (MMP) analysis and descriptor importance analysis were performed to extract transformation rules and give some explanations related to log  D 7.4 . In conclusion, we believe that the consensus model developed in this study can be used as a reliable and promising tool to evaluate log  D 7.4 in drug discovery.

Background: Several randomized controlled trials comparing platelet-rich plasma to standard wound care for chronic wounds have been published. Previous articles have revealed that the role of platelet-rich plasma in promoting chronic wound healing is uncertain. This quantitative meta-analysis was conducted to evaluate whether superior outcomes can be obtained by using platelet-rich plasma in nonhealing ulcers compared with traditional wound care., Methods: The PubMed, EMBASE, EBSCO (Cumulative Index to Nursing and Allied Health Literature), and Cochrane databases were searched through November of 2018 for randomized controlled trials comparing platelet-rich plasma to standard wound care for chronic wounds. For binary outcome measures, we calculated the risk ratio. The continuous outcomes were expressed as the mean differences. Subgroup analyses were also performed according to the type of chronic ulcer., Results: Overall, 630 adult patients in 15 randomized controlled trials from 2000 to 2018 were included. The number of ulcers healed in the platelet-rich plasma group was higher than in the control group, and the difference was statistically significant (risk ratio, 1.26; 95 percent CI, 1.11 to 1.42; p = 0.0003). During the fourth week of follow-up, the number of ulcers healed in the platelet-rich plasma group was significantly greater than in the control group (risk ratio, 3.50; 95 percent CI, 1.80 to 6.81; p = 0.0002)., Conclusions: Platelet-rich plasma is a valuable and safe treatment dressing for chronic nonhealing ulcers; it is simple to prepare and has remarkable effects. Further high-quality prospective studies are necessary to validate these results., Clinical Question/level of Evidence: Therapeutic, II.

Background: Infantile hemangioma (IH) is the most common childhood benign vascular tumor. Recently, propranolol has been found to be an effective therapy for IH, but its mechanism of action is not yet understood. Hemangioma stem cells (HemSCs) have a mesenchymal morphology, robust proliferation, and multilineage differentiation (into adipocytes). Therefore, we hypothesized that propranolol could accelerate the transdifferentiation of HemSCs and prevent the growth of proliferating IH. In this study, the fibrofatty tissue of IH that received therapy with propranolol appeared much earlier than without the treatment., Methods: We isolated HemSCs with CD133-tagged immunomagnetic beads, and then we used flow cytometry technology to analyze the HemSC phenotypes and determine whether propranolol induced HemSC death. The proliferation and adipogenesis abilities of propranolol-treated HemSCs were analyzed by 3-(4,5)-dimethylthiahiazo(-z-y1)-3,5-diphenytetrazoliumromide assay, Oil red O staining, and Western blotting., Results: We observed that the HemSC morphological traits gradually became spindle shaped, like fibroblasts, and the average extraction yield of HemSCs was about 0.25%. The HemSCs had high rates of expression for CD90 (98.8%) and CD105 (97.8%) but did not significantly express CD31 (0.7%). We also found a 100 μM concentration of propranolol cutoff point. Propranolol did not affect HemSC survival significantly at low concentrations (6.25, 12.5, 25.0, and 50.0 μM). However, propranolol resulted in a sharp and significant variation in cell morphology and survival rates at high concentrations (100, 200, and 400 μM). The results suggest that treatment with propranolol inhibited HemSC proliferation and induced cell death and apoptosis in a concentration-dependent manner. Oil droplets determined by Oil red O staining showed that propranolol increased the transdifferentiation rate of HemSCs into adipocytes. Furthermore, the expressions of phosphorylated AKT and peroxisome proliferator-activated receptor gamma (PPARγ) were increased with a 100 μM concentration of propranolol in HemSC culture., Conclusions: Our study found that propranolol inhibited proliferation, induced apoptosis and necrosis, and promoted differentiation of HemSCs. Propranolol may upregulate PPARγ via PI3K pathways, thereby accelerating lipogenesis and enhancing IH HemSC adipogenesis.

Aggregation has been posing a great challenge in drug discovery. Current computational approaches aiming to filter out aggregated molecules based on their similarity to known aggregators, such as Aggregator Advisor, have low prediction accuracy, and therefore development of reliable in silico models to detect aggregators is highly desirable. In this study, we built a data set consisting of 12 119 aggregators and 24 172 drugs or drug candidates and then developed a group of classification models based on the combination of two ensemble learning approaches and five types of molecular representations. The best model yielded an accuracy of 0.950 and an area under the curve (AUC) value of 0.987 for the training set, and an accuracy of 0.937 and an AUC of 0.976 for the test set. The best model also gave reliable predictions to the external validation set with 5681 aggregators since 80% of molecules were predicted to be aggregators with a prediction probability higher than 0.9. More importantly, we explored the relationship between colloidal aggregation and molecular features, and generalized a set of simple rules to detect aggregators. Molecular features, such as log D , the number of hydroxyl groups, the number of aromatic carbons attached to a hydrogen atom, and the number of sulfur atoms in aromatic heterocycles, would be helpful to distinguish aggregators from nonaggregators. A comparison with numerous existing druglikeness and aggregation filtering rules and models used in virtual screening verified the high reliability of the model and rules proposed in this study. We also used the model to screen several curated chemical databases, and almost 20% of molecules in the evaluated databases were predicted as aggregators, highlighting the potential high risk of aggregation in screening. Finally, we developed an online Web server of ChemAGG ( http://admet.scbdd.com/ChemAGG/index ), which offers a freely available tool to detect aggregators.