ChemDes: an integrated web-based platform for molecular descriptor and fingerprint computation

Exposure experiments and machine learning revealed that personal care products can significantly increase transdermal exposure of SVOCs from the environment

We investigated the impacts of personal care products (PCPs) on dermal exposure to semi-volatile organic compounds (SVOCs), including phthalates, organophosphate esters, polycyclic aromatic hydrocarbons (PAHs), ultraviolet filters, and p-phenylenediamines, through an experiment from volunteers, explored the impact mechanisms of PCP ingredients on dermal exposure, and predicted the PCP effects on SVOC concentrations in human serum using machine learning. After applying PCPs, namely lotion, baby oil, sunscreen, and blemish balm, the dermal adsorption of SVOCs increased significantly by 1.63 ± 0.62, 1.97 ± 0.73, 1.91 ± 0.48, and 2.03 ± 0.59 times, respectively, probably due to the absorption effects of PCP ingredients. Ingredient tocopherol can increase dermal adsorption of SVOCs by 2.59 ± 1.60 times. PCPs can either increase or decrease the SVOC transdermal exposure risks, depending on the properties of their ingredients. Blemish balm caused the highest hazard quotient for certain SVOCs, while tris(2-chloroethyl) phosphate (TCEP) exhibited the highest hazard quotient. We predicted the SVOC concentrations in serum before and after applying PCPs based on the PCP-increased skin permeation doses and machine learning. PCPs can significantly increase the serum concentrations of PAHs with 2-3 rings and TCEP. This study first revealed that PCPs can significantly increase the dermal exposure of SVOCs from the surroundings, resulting in potentially higher health risks.

Explainable machine learning models enhance prediction of PFAS bioactivity using quantitative molecular surface analysis-derived representation

The extensive use of per- and polyfluoroalkyl substances (PFAS) in industrial and consumer products poses health risks due to their toxicity. Computational toxicology approaches, particularly quantitative structure-activity relationship (QSAR) models are essential for predicting PFAS bioactivity. However, established QSAR models including machine learning-based ones with traditional molecular descriptors such as constitutional, topological, and geometric descriptors, have limited predictive capability and interpretability. Herein, we proposed a novel machine learning approach that leverages quantitative molecular surface analysis (QMSA) of molecular electrostatic potential. Using QMSA descriptors, five machine learning models (e.g., random forest) achieved outstanding performance, with best accuracy of 0.950 ± 0.017, AUC-ROC of 0.938 ± 0.012, F1-score of 0.734 ± 0.024, and MCC of 0.684 ± 0.111 for five targets (tyrosyl-DNA phosphodiesterase 1 in the absence/presence of camptothecin, ATXN2 protein, transcription factor SMAD3, and transcription factor NRF2), which outperform previously reported models. SHAP analyses revealed that estimated density, molecular volume, positive surface area, and nonpolar surface area were the most important descriptors. These descriptors were deeply involved in PFAS binding to target proteins via non-covalent interactions as evidenced by molecular docking and molecular dynamics simulations. Our results demonstrated that QMSA descriptors-based machine learning models are capable of predicting PFAS toxicity with extraordinary performance and interpretability. This study provides a novel machine learning framework for the high-throughput and cost-effective screening of high-risk emerging PFAS in aquatic environments. By identifying the contaminants that should be prioritized for regulation and treatment among the growing number of PFAS, our work aids in water quality monitoring and risk assessment, and guides decision-making in aquatic environmental management. Furthermore, this work enhances our understanding of the molecular mechanisms involved in PFAS bioactivity.

Identifying traditional Chinese medicine combinations for breast cancer treatment based on transcriptional regulation and chemical structure

Breast cancer (BC) is a prevalent form of cancer among women. Despite the emergence of numerous therapies over the past few decades, few have achieved the ideal therapeutic effect due to the heterogeneity of BC. Drug combination therapy is seen as a promising approach to cancer treatment. Traditional Chinese medicine (TCM), known for its multicomponent nature, has been validated for its anticancer properties, likely due to the synergy effect of the key components. However, identifying effective component combinations from TCM is challenging due to the vast combination possibilities and limited prior knowledge. This study aims to present a strategy for discovering synergistic compounds based on transcriptional regulation and chemical structure. First, BC-related gene sets were used to screen TCM-derived compound combinations guided by synergistic regulation. Then, machine learning models incorporating chemical structural features were established to identify potential compound combinations. Subsequently, the pair of honokiol and neochlorogenic acid was selected by integrating the results of compound combination screening. Finally, cell experiments were conducted to confirm the synergistic effect of the pair against BC. Overall, this study offers an integrated screening strategy to discover compound combinations of TCM against BC. The tumor cell suppression effect of the honokiol and neochlorogenic acid pair validated the effectiveness of the proposed strategy.

Enhancing lipid identification in LC-HRMS data through machine learning-based retention time prediction

The comprehensive identification of peaks in untargeted lipidomics using LC-MS/MS remains a significant challenge. Confidence in lipid annotation can be greatly improved by integrating a highly accurate machine learning-based retention time prediction model. Such an approach enables the identification of lipids for understanding pathogenic mechanisms, biomarker discovery, and drug screening. In this study, we developed a machine learning model to predict retention times and facilitate lipid peak annotations in LC-MS-based untargeted lipidomics. Our model achieved high correlation coefficients of 0.998 and 0.990, with mean absolute errors (MAE) of 0.107 min and 0.240 min for the training and test sets, respectively. External validation showed similarly strong performance, with correlations of 0.991 and 0.978, and MAE values of 0.241 min and 0.270 min. We also compared the impact of molecular descriptors and molecular fingerprints on the model's performance, finding that molecular descriptors outperformed molecular fingerprints across all datasets when using Random Forest (RF) for model construction. Notably, this retention time calibration model demonstrates robust performance across chromatographic systems with comparable gradients and flow rates. Overall, this machine learning model enhances lipid annotation accuracy and reduces errors in untargeted lipidomics, improving data analysis across multiple datasets.

Rational design of potent phosphopeptide binders to endocrine Snk PBD domain by integrating machine learning optimization, molecular dynamics simulation, binding energetics rescoring, and in vitro affinity assay

Human Snk is an evolutionarily conserved serine/threonine kinase essential for the maintenance of endocrine stability. The protein consists of a N-terminal catalytic domain and a C-terminal polo-box domain (PBD) that determines subcellular localization and substrate specificity. Here, an integrated strategy is described to explore the vast structural diversity space of Snk PBD-binding phosphopeptides at a molecular level using machine learning modeling, annealing optimization, dynamics simulation, and energetics rescoring, focusing on the recognition specificity and motif preference of the Snk PBD domain. We further performed a systematic rational design of potent phosphopeptide ligands for the domain based on the harvested knowledge, from which a few potent binders were also confirmed by fluorescence-based assays. A phosphopeptide PP17 was designed as a good binder with affinity improvement by 6.7-fold relative to the control PP0, while the other three designed phosphopeptides PP7, PP13, and PP15 exhibit a comparable potency with PP0. In addition, a basic recognition motif that divides potent Snk PBD-binding sequences into four residue blocks was defined, namely [Χ-5Χ-4]block1-[Ω-3Ω-2Ω-1]block2-[pS0/pT0]block3-[Ψ+1]block4, where the X represents any amino acid, Ω indicates polar amino acid, Ψ denotes hydrophobic amino acid, and pS0/pT0 is the anchor phosphoserine/phosphothreonine at reference residue position 0.

ToxinPredictor: Computational models to predict the toxicity of molecules

Predicting the toxicity of molecules is essential in fields like drug discovery, environmental protection, and industrial chemical management. While traditional experimental methods are time-consuming and costly, computational models offer an efficient alternative. In this study, we introduce ToxinPredictor, a machine learning-based model to predict the toxicity of small molecules using their structural properties. The model was trained on a curated dataset of 7550 toxic and 6514 non-toxic molecules, leveraging feature selection techniques like Boruta and PCA. The best-performing model, a Support Vector Machine (SVM), achieved state-of-the-art results with an AUROC of 91.7%, F1-score of 84.9%, and accuracy of 85.4%, outperforming existing solutions. SHAP analysis was applied to the SVM model to identify the most important molecular descriptors contributing to toxicity predictions, enhancing interpretability. Despite challenges related to data quality, ToxinPredictor provides a reliable framework for toxicity risk assessment, paving the way for safer drug development and improved environmental health assessments. We also created a user-friendly webserver, ToxinPredictor (https://cosylab.iiitd.edu.in/toxinpredictor) to facilitate the search and prediction of toxic compounds.

InterDIA: Interpretable prediction of drug-induced autoimmunity through ensemble machine learning approaches

Drug-induced autoimmunity (DIA) is a non-IgE immune-related adverse drug reaction that poses substantial challenges in predictive toxicology due to its idiosyncratic nature, complex pathogenesis, and diverse clinical manifestations. To address these challenges, we developed InterDIA, an interpretable machine learning framework for predicting DIA toxicity based on molecular physicochemical properties. Multi-strategy feature selection and advanced ensemble resampling approaches were integrated to enhance prediction accuracy and overcome data imbalance. The optimized Easy Ensemble Classifier achieved robust performance in both 10-fold cross-validation (AUC value of 0.8836 and accuracy of 82.81 %) and external validation (AUC value of 0.8930 and accuracy of 85.00 %). Paired case studies of hydralazine/phthalazine and procainamide/N-acetylprocainamide demonstrated the model's capacity to discriminate between structurally similar compounds with distinct immunogenic potentials. Mechanistic interpretation through SHAP (SHapley Additive exPlanations) analysis revealed critical physicochemical determinants of DIA, including molecular lipophilicity, partial charge distribution, electronic states, polarizability, and topological features. These molecular signatures were mechanistically linked to key processes in DIA pathogenesis, such as membrane permeability and tissue distribution, metabolic bioactivation susceptibility, immune protein recognition and binding specificity. SHAP dependence plots analysis identified specific threshold values for key molecular features, providing novel insights into structure-toxicity relationships in DIA. To facilitate practical application, we developed an open-access web platform enabling batch prediction with real-time visualization of molecular feature contributions through SHAP waterfall plots. This integrated framework not only advances our mechanistic understanding of DIA pathogenesis from a molecular perspective but also provides a valuable tool for early assessment of autoimmune toxicity risk during drug development.

Assessing chemical exposure risk in breastfeeding infants: An explainable machine learning model for human milk transfer prediction

Breast milk is essential for infant health, but the transfer of xenobiotic chemicals poses significant risks. Ethical challenges in clinical trials necessitate the use of in vitro predictive models to assess chemical exposure risks in breastfeeding infants. This study introduces an explainable machine learning model to predict the risk of chemical transfer through human milk. Our novel framework integrates ensemble resampling methods with advanced feature selection techniques, addressing data imbalance and enhancing predictive accuracy. The balanced random forest classifier, optimized using the genetic algorithm for feature selection, achieved an area under the receiver operating characteristic curve (AUC) of 0.8708 and an accuracy of 82.67 % on the internal test set, with an accuracy of 86.36 % on the external validation set. The integration of the SHapley Additive exPlanations approach provided deeper insights by revealing how specific chemical properties influence the transfer of high-risk compounds into breast milk. This enhanced interpretability offers a clearer understanding of the associated risks and informs strategies for their mitigation. Structural alert analysis further identified molecular fragments linked to high-risk chemicals, enabling targeted risk assessments. Additionally, the model was applied to evaluate the transfer risks of FDA-approved drugs from 2019 to 2024, identifying several with high transfer probabilities. To broaden its application, we developed an online prediction tool that offers real-time risk assessments, providing an accessible resource for healthcare professionals and researchers. These contributions present a robust, ethically sound tool for assessing chemical exposure risks in breastfeeding infants, supporting informed decisions on drug use and environmental contaminant exposure.

ML-based Models as a Strategy to Discover Novel Antiepileptic Drugs Targeting Sodium Receptor Channel

BACKGROUND

Epilepsy remains the most common and chronic disorder demanding longterm management. The impact of epilepsy disease is a cause of great concern and has resulted in efforts to develop treatment for epilepsy. It occurs due to an increase in neuronal excitability produced by changes affecting the voltage-dependent properties of Voltage-gated Sodium Channels (VGSCs).

MATERIALS AND METHODS

Weka, a popular suite for machine learning techniques, was used on a dataset comprising 1781 chemical compounds, showing inhibition activity for sodium channel protein IX alpha subunit. After the analysis of the dataset obtained from ChEMBL, molecular fingerprints were computed for the molecules by the ChemDes server. Different classifiers available in the Weka software were explored to find out the algorithm that could be more suitable for the dataset or produce the highest accuracy for the classification of molecules as active or inactive.

RESULTS

In this work, a comprehensive comparison of different classifiers in the Weka suite for the prediction of active, inactive, and intermediate classes of molecules showing inhibition against human NaV1.7 protein was made. The prediction accuracy of these classifiers was assessed based on performance measures, including accuracy, Root Mean Squared Error (RMSE), Receiver Operating Characteristic (ROC), precision, Mathews Correlation Coefficient (MCC), recall, and Fmeasure. The comparison of results for model performance demonstrated that the OneR classifier performed best over others when validated using percentage split, cross-validation, and supplied test methods. J48 and Bagging also performed equally well in the prediction of different classes with an MCC value of 1, ROC area equal to 1, and RMSE close to 0.

CONCLUSION

Machine Learning (ML) tools provide a fast, reliable, and cost-effective approach required to identify or predict inhibitory molecules for the treatment of a disease. This study shows that the ML methods, particularly OneR, J48, and Bagging have the ability to identify active and inactive classes of compounds for the human NaV1.7 protein target. Such predictive models may provide a reliable and time-saving approach that can aid in the design of potential inhibitors for the treatment of epilepsy disease.

Antilipase and antioxidant activities of topiramate-phenolic acid conjugates: Computational modelling, synthesis, and in-vitro investigations

A series of ten topiramate-phenolic acid conjugates (T1-T10) were synthesized, and evaluated for their pancreatic lipase inhibitory and antioxidant potentials. The design of the compounds reflected the structural attributes extracted from robust QSAR models developed for predicting the pancreatic lipase inhibition potency. Conjugate T4 competitively inhibited pancreatic lipase with IC50 value of 8.96 μM, comparable to the standard drug, orlistat (IC50 = 11.68 μM). Molecular docking of T4 into the active site of human PL (PDB ID: 1LPB) revealed strong binding score of -11.54 kcal/mol. Molecular dynamics simulation of T4 complexed with pancreatic lipase, confirmed the role of phenolic acid core in stabilizing the ligand through hydrophobic interactions (maximum observed RMSD = 3.77 Å). Additionally, T4 with its LUMO (-0.20254) and HOMO (0.30502) values, abstracted from DFT studies, depicts considerable promise in the pursuit of selecting an effective enzyme inhibitor binding to the enzyme's active site and disrupting its catalytic function. The conjugation of topiramate with phenolic acids has imparted potential antioxidant properties to the synthesized conjugates especially T3, T4 and T5. Conclusively, with good safety profile as predicted from in silico ADMET studies, potent pancreatic lipase inhibition and free radical quenching, T4 stands taller as promising anti-obesity drug candidate.

Molecular modelling studies of substituted indole derivatives as novel influenza a virus inhibitors

The emergence of drug-resistant strains motivate researchers to find new innovative anti-IAV candidates with a different mode of action. In this work, molecular modelling strategies, such as 2D-QSAR, 3D-QSAR, molecular docking, molecular dynamics, FMOs, and ADMET were applied to some substituted indoles as IAV inhibitors. The best-developed 2D-QSAR models, MLR (Q2 = 0.7634, R2train = 0.8666) and ANN[4-3-1] (Q2 = 0.8699, R2train = 0.8705) revealed good statistical validation for the inhibitory response predictions. The 3D-QSAR models, CoMFA (Q2 = 0.504, R2train = 0.805) and CoMSIA/SEDHA (Q2 = 0.619, R2train = 0.813) are selected as the best 3D models following the global thresholds. In addition, the contour maps generated from the CoMFA and CoMSIA models illustrate the relationship between the molecular fields and the inhibitory effects of the studied molecules. The results of the studies led to the design of five new molecules (24a-e) with enhanced anti-IAV activities and binding potentials using the most active molecule (24) as the template scaffold. The conformational stability of the best-designed molecules with the NA protein showed hydrophobic and H-bonds with the key residues from the molecular dynamics simulations of 100 ns. Furthermore, the global reactivity indices from the DFT calculations portrayed the relevance of 24c in view of its smaller band gap as also justified by our QSAR and molecular simulation studies.Communicated by Ramaswamy H. Sarma.

In silico design of dehydrophenylalanine containing peptide activators of glucokinase using pharmacophore modelling, molecular dynamics and machine learning: implications in type 2 diabetes

Diabetes represents a significant global health challenge associated with substantial healthcare costs and therapeutic complexities. Current diabetes therapies often entail adverse effects, necessitating the exploration of novel agents. Glucokinase (GK), a key enzyme in glucose homeostasis, primarily regulates blood glucose levels in hepatocytes and pancreatic cells. Unlike other hexokinases, GK exhibits unique kinetic properties, such as a high Km and lack of feedback inhibition, allowing it to function as a glucose sensor Glucokinase activators (GKAs) have emerged as promising candidates for managing type-2 diabetes by allosterically enhancing GK activity. Despite initial promise, existing GKAs face significant safety concerns, driving the need for compounds with improved safety profiles. This study introduces a novel chemical scaffold within the GKA landscape: peptide-based GKAs incorporating non-standard amino acid residues such as α,β-dehydrophenylalanine (ΔPhe/ΔF). A virtual library containing 3,368,000 peptides was constructed and screened using a hybrid pharmacophore, namely DHRR (D: donor; H: hydrogen; R: aromatic ring). Molecular docking and molecular dynamics simulations assisted in identifying three peptides, Pep-11, Pep-15, and Pep-16, which depicted stable binding at the allosteric site of Glucokinase. These peptides were synthesized using a combination of solid and solution phase synthesis methods. In vitro enzymatic activity of glucokinase was increased by at least 1.5 times in the presence of these peptides. Several machine learning algorithms were explored as alternatives to conventional in-silico methods for predicting GK activity. Regression and tree-based algorithms outperformed other methods, with the logistic regression and random forest classifiers both achieving an ROC-AUC of 0.98.

Multi-method computational evaluation of the inhibitors against leucine-rich repeat kinase 2 G2019S mutant for Parkinson's disease

Leucine-rich repeat kinase 2 G2019S mutant (LRRK2 G2019S) is a potential target for Parkinson's disease therapy. In this work, the computational evaluation of the LRRK2 G2019S inhibitors was conducted via a combined approach which contains a preliminary screening of a large database of compounds via similarity and pharmacophore, a secondary selection via structure-based affinity prediction and molecular docking, and a rescoring treatment for the final selection. MD simulations and MM/GBSA calculations were performed to check the agreement between different prediction methods for these inhibitors. 331 experimental ligands were collected, and 170 were used to build the structure-activity relationship. Eight representative ligand structural models were employed in similarity searching and pharmacophore screening over 14 million compounds. The process for selecting proper molecular descriptors provides a successful sample which can be used as a general strategy in QSAR modelling. The rescoring used in this work presents an alternative useful treatment for ranking and selection.

A Semi-Quantitative Approach to Nontarget Compositional Analysis of Complex Samples

Nontarget analysis (NTA) by liquid chromatography coupled to high-resolution mass spectrometry improves the capacity to comprehend the molecular composition of complex mixtures compared to targeted analysis techniques. However, the detection of unknown compounds means that quantification in NTA is challenging. This study proposes a new semi-quantitative methodology for use in the NTA of organic aerosol. Quantification of unknowns is achieved using the average ionization efficiency of multiple quantification standards which elute within the same retention time window as the unknown analytes. In total, 110 authentic standards constructed 25 retention time windows for the quantification of oxygenated (CHO) and organonitrogen (CHON) species. The method was validated on extracts of biomass burning organic aerosol (BBOA) and compared to quantification with authentic standards and had an average prediction error of 1.52 times. Furthermore, 70% of concentrations were estimated within a factor of 2 (prediction errors between 0.5 and 2 times) from the authentic standard quantification. The semi-quantification method also showed good agreement for the quantification of CHO compounds compared to predictive ionization efficiency-based methods, whereas for CHON species, the prediction error of the semi-quantification method (1.63) was significantly lower than the predictive ionization efficiency approach (14.94). Application to BBOA for the derivation of relative abundances of CHO and CHON species showed that using peak area underestimated the relative abundance of CHO by 19% and overestimated that of CHON by 11% compared to the semi-quantification method. These differences could lead to significant misinterpretations of source apportionment in complex samples, highlighting the need to account for ionization differences in NTA approaches.

Machine learning models for predicting the rejection of organic pollutants by forward osmosis and reverse osmosis membranes and unveiling the rejection mechanisms

While forward osmosis (FO) and reverse osmosis (RO) processes have been proven effective in rejecting organic pollutants, the rejection rate is highly dependent on compound and membrane characteristics, as well as operating conditions. This study aims to establish machine learning (ML) models for predicting the rejection of organic pollutants by FO and RO and providing insights into the underlying rejection mechanisms. Among the 14 ML models established, the random forest model (R2 = 0.85) and extreme gradient boosting model (R2 = 0.92) emerged as the best-performing models for FO and RO, respectively. Shapley additive explanations (SHAP) analysis identified the length of the compound, water flux, and hydrophobicity as the top three variables contributing to the FO model. For RO, in addition to the length of the compound and operating pressure, advanced variables including four molecular descriptors (e.g., ATSC2m and Balaban J) and three fingerprints (e.g., C=C double bond and carbonyl group) significantly contributed to the prediction. Besides, the associations between these highly ranked variables and their SHAP values shed light on the rejection mechanisms, such as size exclusion, adsorption, hydrophobic interaction, and electrostatic interaction, and illustrate the role of the operating parameters, such as the FO permeate water flux and RO operating pressure, in the rejection process. These findings provide interpretable predictive models for the removal of organic pollutants and advance the mechanistic understanding of the rejection mechanisms in the FO and RO processes.

Advancing Breathomics through Accurate Discrimination of Endogenous from Exogenous Volatiles in Breath

Breathomics, a growing field in exposure monitoring and clinical diagnostics, has faced accuracy challenges due to unclear contributing factors. This study aims to enhance the potential of breathomics in various frontiers by categorizing exhaled volatile organic compounds (VOCs) as endogenous or exogenous. Analyzing ambient air and breath samples from 271 volunteers via TD-GC × GC-TOF MS/FID, we identify and quantify 50 common VOCs in exhaled breath. Advanced quantitative structure-property relationships and compartment models are employed to obtain VOCs kinetic parameters. This in-depth approach allows us to accurately determine the alveolar concentration of VOCs and further discern their origins, facilitating personalized application of breathomics in exposure assessment and disease diagnosis. Our findings demonstrate that prolonged external exposure turns humans into secondary pollutant sources. Analysis of endogenous VOCs reveals that internal exposure poses more significant health risks than external. Moreover, by correcting environmental backgrounds, we improve the accuracy of gastrointestinal disease diagnostic models by 15-25%. This advancement in identifying VOC origins via compartmental models promises to elevate the clinical relevance of breathomics, marking a leap forward in exposure assessment and precision medicine.

Machine Learning Models for Predicting Bioavailability of Traditional and Emerging Aromatic Contaminants in Plant Roots

To predict the behavior of aromatic contaminants (ACs) in complex soil-plant systems, this study developed machine learning (ML) models to estimate the root concentration factor (RCF) of both traditional (e.g., polycyclic aromatic hydrocarbons, polychlorinated biphenyls) and emerging ACs (e.g., phthalate acid esters, aryl organophosphate esters). Four ML algorithms were employed, trained on a unified RCF dataset comprising 878 data points, covering 6 features of soil-plant cultivation systems and 98 molecular descriptors of 55 chemicals, including 29 emerging ACs. The gradient-boosted regression tree (GBRT) model demonstrated strong predictive performance, with a coefficient of determination (R2) of 0.75, a mean absolute error (MAE) of 0.11, and a root mean square error (RMSE) of 0.22, as validated by five-fold cross-validation. Multiple explanatory analyses highlighted the significance of soil organic matter (SOM), plant protein and lipid content, exposure time, and molecular descriptors related to electronegativity distribution pattern (GATS8e) and double-ring structure (fr_bicyclic). An increase in SOM was found to decrease the overall RCF, while other variables showed strong correlations within specific ranges. This GBRT model provides an important tool for assessing the environmental behaviors of ACs in soil-plant systems, thereby supporting further investigations into their ecological and human exposure risks.

Prediction of adverse drug reactions using demographic and non-clinical drug characteristics in FAERS data

The presence of adverse drug reactions (ADRs) is an ongoing public health concern. While traditional methods to discover ADRs are very costly and limited, it is prudent to predict ADRs through non-invasive methods such as machine learning based on existing data. Although various studies exist regarding ADR prediction using non-clinical data, a process that leverages both demographic and non-clinical data for ADR prediction is missing. In addition, the importance of individual features in ADR prediction has yet to be fully explored. This study aims to develop an ADR prediction model based on demographic and non-clinical data, where we identify the highest contributing factors. We focus our efforts on 30 common and severe ADRs reported to the Food and Drug Administration (FDA) between 2012 and 2023. We have developed a random forest (RF) and deep learning (DL) machine learning model that ingests demographic data (e.g., Age and Gender of patients) and non-clinical data, which includes chemical, molecular, and biological drug characteristics. We successfully unified both demographic and non-clinical data sources within a complete dataset regarding ADR prediction. Model performances were assessed via the area under the receiver operating characteristic curve (AUC) and the mean average precision (MAP). We demonstrated that our parsimonious models, which include only the top 20 most important features comprising 5 demographic features and 15 non-clinical features (13 molecular and 2 biological), achieve ADR prediction performance comparable to a less practical, feature-rich model consisting of all 2,315 features. Specifically, our models achieved an AUC of 0.611 and 0.674 for RF and DL algorithms, respectively. We hope our research provides researchers and clinicians with valuable insights and facilitates future research designs by identifying top ADR predictors (including demographic information) and practical parsimonious models.

Molecular similarity in chemical informatics and predictive toxicity modeling: from quantitative read-across (q-RA) to quantitative read-across structure-activity relationship (q-RASAR) with the application of machine learning

This article aims to provide a comprehensive critical, yet readable, review of general interest to the chemistry community on molecular similarity as applied to chemical informatics and predictive modeling with a special focus on read-across (RA) and read-across structure-activity relationships (RASAR). Molecular similarity-based computational tools, such as quantitative structure-activity relationships (QSARs) and RA, are routinely used to fill the data gaps for a wide range of properties including toxicity endpoints for regulatory purposes. This review will explore the background of RA starting from how structural information has been used through to how other similarity contexts such as physicochemical, absorption, distribution, metabolism, and elimination (ADME) properties, and biological aspects are being characterized. More recent developments of RA's integration with QSAR have resulted in the emergence of novel models such as ToxRead, generalized read-across (GenRA), and quantitative RASAR (q-RASAR). Conventional QSAR techniques have been excluded from this review except where necessary for context.

Polypharmacology prediction: the long road toward comprehensively anticipating small-molecule selectivity to de-risk drug discovery

INTRODUCTION

Small molecules often bind to multiple targets, a behavior termed polypharmacology. Anticipating polypharmacology is essential for drug discovery since unknown off-targets can modulate safety and efficacy - profoundly affecting drug discovery success. Unfortunately, experimental methods to assess selectivity present significant limitations and drugs still fail in the clinic due to unanticipated off-targets. Computational methods are a cost-effective, complementary approach to predict polypharmacology.

AREAS COVERED

This review aims to provide a comprehensive overview of the state of polypharmacology prediction and discuss its strengths and limitations, covering both classical cheminformatics methods and bioinformatic approaches. The authors review available data sources, paying close attention to their different coverage. The authors then discuss major algorithms grouped by the types of data that they exploit using selected examples.

EXPERT OPINION

Polypharmacology prediction has made impressive progress over the last decades and contributed to identify many off-targets. However, data incompleteness currently limits most approaches to comprehensively predict selectivity. Moreover, our limited agreement on model assessment challenges the identification of the best algorithms - which at present show modest performance in prospective real-world applications. Despite these limitations, the exponential increase of multidisciplinary Big Data and AI hold much potential to better polypharmacology prediction and de-risk drug discovery.

AI-Powered Mining of Highly Customized and Superior ESIPT-Based Fluorescent Probes

Excited-state intramolecular proton transfer (ESIPT) has attracted great attention in fluorescent sensors and luminescent materials due to its unique photobiological and photochemical features. However, the current structures are far from meeting the specific demands for ESIPT molecules in different scenarios; the try-and-error development method is labor-intensive and costly. Therefore, it is imperative to devise novel approaches for the exploration of promising ESIPT fluorophores. This research proposes an artificial intelligence approach aiming at exploring ESIPT molecules efficiently. The first high-quality ESIPT dataset and a multi-level prediction system are constructed that realized accurate identification of ESIPT molecules from a large number of compounds under a stepwise distinguishing from conventional molecules to fluorescent molecules and then to ESIPT molecules. Furthermore, key structural features that contributed to ESIPT are revealed by using the SHapley Additive exPlanations (SHAP) method. Then three strategies are proposed to ensure the ESIPT process while keeping good safety, pharmacokinetic properties, and novel structures. With these strategies, >700 previously unreported ESIPT molecules are screened from a large pool of 570 000 compounds. The ESIPT process and biosafety of optimal molecules are successfully validated by quantitative calculation and experiment. This novel approach is expected to bring a new paradigm for exploring ideal ESIPT molecules.

Interpretable AI and Machine Learning Classification for Identifying High-Efficiency Donor-Acceptor Pairs in Organic Solar Cells

To enhance the efficiency of organic solar cells, accurately predicting the efficiency of new pairs of donor and acceptor materials is crucial. Presently, most machine learning studies rely on regression models, which often struggle to establish clear rules for distinguishing between high- and low-performing donor-acceptor pairs. This study proposes a novel approach by integrating interpretable AI, specifically using Shapely values, with four supervised machine learning classification models, namely, support vector machines, decision trees, random forest, and gradient boosting. These models aim to identify high-efficiency donor-acceptor pairs based solely on chemical structures and to extract important features that establish general design principles for distinguishing between high- and low-efficiency pairs. For validation purposes, an unsupervised machine learning algorithm utilizing loading vectors obtained from the principal component analysis is employed to identify crucial features associated with high-efficiency donor-acceptor pairs. Interestingly, the features identified by the supervised machine learning approach were found to be a subset of those identified by the unsupervised method. Noteworthy features include the van der Waals surface area, partial equalization of orbital electronegativity, Moreau-Broto autocorrelation, and molecular substructures. Leveraging these features, a backward-working model can be developed, facilitating exploration across a wide array of materials used in organic solar cells. This innovative approach will help navigate the vast chemical compound space of donor and acceptor materials essential in creating high-efficiency organic solar cells.

Accurate Prediction of Rat Acute Oral Toxicity and Reference Dose for Thousands of Polycyclic Aromatic Hydrocarbon Derivatives Based on Chemometric QSAR and Machine Learning

Acute oral toxicity is currently not available for most polycyclic aromatic hydrocarbons (PAHs), especially their derivatives, because it is cost-prohibitive to experimentally determine all of them. Here, quantitative structure-activity relationship (QSAR) models using machine learning (ML) for predicting the toxicity of PAH derivatives were developed, based on oral toxicity data points of 788 individual substances of rats. Both the individual ML algorithm gradient boosting regression trees (GBRT) and the stacking ML algorithm (extreme gradient boosting + GBRT + random forest regression) provided the best prediction results with satisfactory determination coefficients for both cross-validation and the test set. It was found that those PAH derivatives with fewer polar hydrogens, more large-sized atoms, more branches, and lower polarizability have higher toxicity. Software based on the optimal ML-QSAR model was successfully developed to expand the application potential of the developed model, obtaining reliable prediction of pLD50 values and reference doses for 6893 external PAH derivatives. Among these chemicals, 472 were identified as moderately or highly toxic; 10 out of them had clear environment detection or use records. The findings provide valuable insights into the toxicity of PAHs and their derivatives, offering a standard platform for effectively evaluating chemical toxicity using ML-QSAR models.

Predicting repurposed drugs targeting the NS3 protease of dengue virus using machine learning-based QSAR, molecular docking, and molecular dynamics simulations

Dengue fever, prevalent in Southeast Asian countries, currently lacks effective pharmaceutical interventions for virus replication control. This study employs a strategy that combines machine learning (ML)-based quantitative-structure-activity relationship (QSAR), molecular docking, and molecular dynamics simulations to discover potential inhibitors of the NS3 protease of the dengue virus. We used nine molecular fingerprints from PaDEL to extract features from the NS3 protease dataset of dengue virus type 2 in the ChEMBL database. Feature selection was achieved through the low variance threshold, F-Score, and recursive feature elimination (RFE) methods. Our investigation employed three ML models - support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost) - for classifier development. Our SVM model, combined with SVM-RFE, had the best accuracy (0.866) and ROC_AUC (0.964) in the testing set. We identified potent inhibitors on the basis of the optimal classifier probabilities and docking binding affinities. SHAP and LIME analyses highlighted the significant molecular fingerprints (e.g. ExtFP69, ExtFP362, ExtFP576) involved in NS3 protease inhibitory activity. Molecular dynamics simulations indicated that amphotericin B exhibited the highest binding energy of -212 kJ/mol and formed a hydrogen bond with the critical residue Ser196. This approach enhances NS3 protease inhibitor identification and expedites the discovery of dengue therapeutics.

Flavonoid as a Potent Antioxidant: Quantitative Structure-Activity Relationship Analysis, Mechanism Study, and Molecular Design by Synergizing Molecular Simulation and Machine Learning

In this work, a quantitative structure-antioxidant activity relationship of flavonoids was performed using a machine learning (ML) method. To achieve lipid-soluble, highly antioxidant flavonoids, 398 molecular structures with various substitute groups were designed based on the flavonoid skeleton. The hydrogen dissociation energies (ΔG1, ΔG2, and ΔG3) related to multiple hydrogen atom transfer processes and the solubility parameter (δ) of flavonoids were calculated using molecular simulation. The group decomposition results and the calculated antioxidant parameters constituted the ML data set. The artificial neural network and random forest models were constructed to predict and analyze the contribution of the substitute groups and positions to the antioxidant activity. The results showed the hydroxyl group at positions B4', B5', and B6' and the branched alkyl group at position C3 in the flavonoid skeleton were the optimal choice for improving antioxidant activity and compatibility with apolar organic materials. Compared to the pyrogallol group-grafted flavonoid, the designed potent flavonoid decreased ΔG1 and δ by 2.2 and 15.1%, respectively, while ΔG2 and ΔG3 kept the favorable lower values. These findings suggest that an efficient flavonoid prefers multiple ortho-phenolic hydroxyl groups and suitable sites with hydrophobic groups. The combination of molecular simulation and the ML method may offer a new research approach for the molecular design of novel antioxidants.

Multinomial classification of NLRP3 inhibitory compounds based on large scale machine learning approaches

The role of NLRP3 inflammasome in innate immunity is newly recognized. The NLRP3 protein is a family of nucleotide-binding and oligomerization domain-like receptors as well as a pyrin domain-containing protein. It has been shown that NLRP3 may contribute to the development and progression of a variety of diseases, such as multiple sclerosis, metabolic disorders, inflammatory bowel disease, and other auto-immune and auto-inflammatory conditions. The use of machine learning methods in pharmaceutical research has been widespread for several decades. An important objective of this study is to apply machine learning approaches for the multinomial classification of NLRP3 inhibitors. However, data imbalances can affect machine learning. Therefore, a synthetic minority oversampling technique (SMOTE) has been developed to increase the sensitivity of classifiers to minority groups. The QSAR modelling was performed using 154 molecules retrieved from the ChEMBL database (version 29). The accuracy of the multiclass classification top six models was found to fall within ranges of 0.99 to 0.86, and log loss ranges of 0.2 to 2.3, respectively. The results showed that the receiver operating characteristic curve (ROC) plot values significantly improved when tuning parameters were adjusted and imbalanced data was handled. Moreover, the results demonstrated that SMOTE offers a significant advantage in handling imbalanced datasets as well as substantial improvements in overall accuracy of machine learning models. The top models were then used to predict data from unseen datasets. In summary, these QSAR classification models exhibited robust statistical results and were interpretable, which strongly supported their use for rapid screening of NLRP3 inhibitors.

Variations of VEGFR2 Chemical Space: Stimulator and Inhibitory Peptides

The kinase pathway plays a crucial role in blood vessel function. Particular attention is paid to VEGFR type 2 angiogenesis and vascular morphogenesis as the tyrosine kinase pathway is preferentially activated. In silico studies were performed on several peptides that affect VEGFR2 in both stimulating and inhibitory ways. This investigation aims to examine the molecular properties of VEGFR2, a molecule primarily involved in the processes of vasculogenesis and angiogenesis. These relationships were defined by the interactions between Vascular Endothelial Growth Factor receptor 2 (VEGFR2) and the structural features of the systems. The chemical space of the inhibitory peptides and stimulators was described using topological and energetic properties. Furthermore, chimeric models of stimulating and inhibitory proteins (for VEGFR2) were computed using the protein system structures. The interaction between the chimeric proteins and VEGFR was computed. The chemical space was further characterized using complex manifolds and high-dimensional data visualization. The results show that a slightly similar chemical area is shared by VEGFR2 and stimulating and inhibitory proteins. On the other hand, the stimulator peptides and the inhibitors have distinct chemical spaces.

A Review of Machine Learning and QSAR/QSPR Predictions for Complexes of Organic Molecules with Cyclodextrins

Cyclodextrins are macrocyclic rings composed of glucose residues. Due to their remarkable structural properties, they can form host-guest inclusion complexes, which is why they are frequently used in the pharmaceutical, cosmetic, and food industries, as well as in environmental and analytical chemistry. This review presents the reports from 2011 to 2023 on the quantitative structure-activity/property relationship (QSAR/QSPR) approach, which is primarily employed to predict the thermodynamic stability of inclusion complexes. This article extensively discusses the significant developments related to the size of available experimental data, the available sets of descriptors, and the machine learning (ML) algorithms used, such as support vector machines, random forests, artificial neural networks, and gradient boosting. As QSAR/QPR analysis only requires molecular structures of guests and experimental values of stability constants, this approach may be particularly useful for predicting these values for complexes with randomly substituted cyclodextrins, as well as for estimating their dependence on pH. This work proposes solutions on how to effectively use this knowledge, which is especially important for researchers who will deal with this topic in the future. This review also presents other applications of ML in relation to CD complexes, including the prediction of physicochemical properties of CD complexes, the development of analytical methods based on complexation with CDs, and the optimisation of experimental conditions for the preparation of the complexes.

Virtual screening and molecular growth guide the design of inhibitors for the influenza virus drug-resistant mutant M2-V27A/S31N

The influenza A virus matrix protein 2 (AM2) protein is a proton-gated, proton-selective ion channel essential for influenza replication that has been identified as an antiviral target. The drug-resistance of the M2-V27A/S31N strain, which has been growing more prevalent in recent years and has the potential to spread globally, prevents current amantadine inhibitors from having the desired impact. In this study, we compiled the most common influenza A virus strains from 2001-2020 from the U.S. National Center for Biotechnology Information database and hypothesized that M2-V27A/S31N would become a common strain. The lead compound ZINC299830590 was screened for M2-V27A/S31N in the ZINC15 database using a pharmacophore model and molecular descriptors. This lead compound was then optimized by molecular growth, which allowed us to identify important amino acid residues and create interactions with them to produce compound 4. Molecular dynamics simulation showed that the complex of compound 4 and M2-V27A/S31N had certain degrees of stability and flexibility. The binding free energy of compound 4 was calculated using the MM/PB(GB)SA method and totaled -106.525 kcal/mol. Finally, physicochemical and pharmacokinetic profiles were predicted using the Absorption, Distribution, Metabolism, Excretion, and Toxicity model, which indicated the good bioavailability of compound 4. These results provide the basis for further in vivo and in vitro studies to demonstrate that compound 4 is a promising drug candidate against M2-V27A/S31N.Communicated by Ramaswamy H. Sarma.

Uncovering PPAR-γ agonists: An integrated computational approach driven by machine learning

Peroxisome proliferator-activated receptor gamma (PPAR-γ) serves as a nuclear receptor with a pivotal function in governing diverse facets of metabolic processes. In diabetes, the prime physiological role of PPAR-γ is to enhance insulin sensitivity and regulate glucose metabolism. Although PPAR-γ agonists such as Thiazolidinediones are effective in addressing diabetes complications, it is vital to be mindful that they are associated with substantial side effects that could potentially give rise to health challenges. The recent surge in the discovery of selective modulators of PPAR-γ inspired us to formulate an integrated computational strategy by leveraging the promising capabilities of both machine learning and in silico drug design approaches. In pursuit of our objectives, the initial stage of our work involved constructing an advanced machine learning classification model, which was trained utilizing chemical information and physicochemical descriptors obtained from known PPAR-γ modulators. The subsequent application of machine learning-based virtual screening, using a library of 31,750 compounds, allowed us to identify 68 compounds having suitable characteristics for further investigation. A total of four compounds were identified and the most favorable configurations were complemented with docking scores ranging from -8.0 to -9.1 kcal/mol. Additionally, the compounds engaged in hydrogen bond interactions with essential conserved residues including His323, Leu330, Phe363, His449 and Tyr473 that describe the ligand binding site. The stability indices investigated herein for instance root-mean-square fluctuations in the backbone atoms indicated higher mobility in the region of orthosteric site in the presence of agonist with the deviation peaks in the range of 0.07-0.69 nm, signifying moderate conformational changes. The deviations at global level revealed that the average values lie in the range of 0.25-0.32 nm. In conclusion, our identified hits particularly, CHEMBL-3185642 and CHEMBL-3554847 presented outstanding results and highlighted the stable conformation within the orthosteric site of PPAR-γ to positively modulate the activity.

Estimation of passive gastrointestinal absorption of new dual DNA gyrase and topoisomerase IV inhibitors using PAMPA and biopartitioning micellar chromatography and quantitative structure-retention relationship analysis

DNA gyrase and topoisomerase IV play significant role in maintaining the correct structure of DNA during replication and they have been identified as validated targets in antibacterial drug discovery. Inadequate pharmacokinetic properties are responsible for many failures during drug discovery and their estimation in the early phase of this process maximizes the chance of getting useful drug candidates. Passive gastrointestinal absorption of a selected group of thirteen dual DNA gyrase and topoisomerase IV inhibitors was estimated using two in vitro tests - parallel artificial membrane permeability assay (PAMPA) and biopartitioning micellar chromatography (BMC). Due to good correlation between obtained results, passive gastrointestinal absorption of remaining ten compounds was estimated using only BMC. With this experimental setup, it was possible to identify compounds with high values of retention factors (k) and highest expected passive gastrointestinal absorption, and compounds with low values of k for which low passive gastrointestinal absorption is predicted. Quantitative structure-retention relationship (QSRR) modelling was performed by creating multiple linear regression (MLR), partial least squares (PLS) and support vector machines (SVM) models. Descriptors with the highest influence on retention factor were identified and their interpretation can be used for the design of new compounds with improved passive gastrointestinal absorption.

Colorectal Cancer Detection via Metabolites and Machine Learning

Today, colorectal cancer (CRC) diagnosis is performed using colonoscopy, which is the current, most effective screening method. However, colonoscopy poses risks of harm to the patient and is an invasive process. Recent research has proven metabolomics as a potential, non-invasive detection method, which can use identified biomarkers to detect potential cancer in a patient's body. The aim of this study is to develop a machine-learning (ML) model based on chemical descriptors that will recognize CRC-associated metabolites. We selected a set of metabolites found as the biomarkers of CRC, confirmed that they participate in cancer-related pathways, and used them for training a machine-learning model for the diagnostics of CRC. Using a set of selective metabolites and random compounds, we developed a range of ML models. The best performing ML model trained on Stage 0-2 CRC metabolite data predicted a metabolite class with 89.55% accuracy. The best performing ML model trained on Stage 3-4 CRC metabolite data predicted a metabolite class with 95.21% accuracy. Lastly, the best-performing ML model trained on Stage 0-4 CRC metabolite data predicted a metabolite class with 93.04% accuracy. These models were then tested on independent datasets, including random and unrelated-disease metabolites. In addition, six pathways related to these CRC metabolites were also distinguished: aminoacyl-tRNA biosynthesis; glyoxylate and dicarboxylate metabolism; glycine, serine, and threonine metabolism; phenylalanine, tyrosine, and tryptophan biosynthesis; arginine biosynthesis; and alanine, aspartate, and glutamate metabolism. Thus, in this research study, we created machine-learning models based on metabolite-related descriptors that may be helpful in developing a non-invasive diagnosis method for CRC.

Predicting FFAR4 agonists using structure-based machine learning approach based on molecular fingerprints

Free Fatty Acid Receptor 4 (FFAR4), a G-protein-coupled receptor, is responsible for triggering intracellular signaling pathways that regulate various physiological processes. FFAR4 agonists are associated with enhancing insulin release and mitigating the atherogenic, obesogenic, pro-carcinogenic, and pro-diabetogenic effects, normally associated with the free fatty acids bound to FFAR4. In this research, molecular structure-based machine-learning techniques were employed to evaluate compounds as potential agonists for FFAR4. Molecular structures were encoded into bit arrays, serving as molecular fingerprints, which were subsequently analyzed using the Bayesian network algorithm to identify patterns for screening the data. The shortlisted hits obtained via machine learning protocols were further validated by Molecular Docking and via ADME and Toxicity predictions. The shortlisted compounds were then subjected to MD Simulations of the membrane-bound FFAR4-ligand complexes for 100 ns each. Molecular analyses, encompassing binding interactions, RMSD, RMSF, RoG, PCA, and FEL, were conducted to scrutinize the protein-ligand complexes at the inter-atomic level. The analyses revealed significant interactions of the shortlisted compounds with the crucial residues of FFAR4 previously documented. FFAR4 as part of the complexes demonstrated consistent RMSDs, ranging from 3.57 to 3.64, with minimal residue fluctuations 5.27 to 6.03 nm, suggesting stable complexes. The gyration values fluctuated between 22.8 to 23.5 nm, indicating structural compactness and orderliness across the studied systems. Additionally, distinct conformational motions were observed in each complex, with energy contours shifting to broader energy basins throughout the simulation, suggesting thermodynamically stable protein-ligand complexes. The two compounds CHEMBL2012662 and CHEMBL64616 are presented as potential FFAR4 agonists, based on these insights and in-depth analyses. Collectively, these findings advance our comprehension of FFAR4's functions and mechanisms, highlighting these compounds as potential FFAR4 agonists worthy of further exploration as innovative treatments for metabolic and immune-related conditions.

Accurate Characterization of Binding Kinetics and Allosteric Mechanisms for the HSP90 Chaperone Inhibitors Using AI-Augmented Integrative Biophysical Studies

The binding kinetics of drugs to their targets are gradually being recognized as a crucial indicator of the efficacy of drugs in vivo, leading to the development of various computational methods for predicting the binding kinetics in recent years. However, compared with the prediction of binding affinity, the underlying structure and dynamic determinants of binding kinetics are more complicated. Efficient and accurate methods for predicting binding kinetics are still lacking. In this study, quantitative structure-kinetics relationship (QSKR) models were developed using 132 inhibitors targeting the ATP binding domain of heat shock protein 90α (HSP90α) to predict the dissociation rate constant (koff), enabling a direct assessment of the drug-target residence time. These models demonstrated good predictive performance, where hydrophobic and hydrogen bond interactions significantly influence the koff prediction. In subsequent applications, our models were used to assist in the discovery of new inhibitors for the N-terminal domain of HSP90α (N-HSP90α), demonstrating predictive capabilities on an experimental validation set with a new scaffold. In X-ray crystallography experiments, the loop-middle conformation of apo N-HSP90α was observed for the first time (previously, the loop-middle conformation had only been observed in holo-N-HSP90α structures). Interestingly, we observed different conformations of apo N-HSP90α simultaneously in an asymmetric unit, which was also observed in a holo-N-HSP90α structure, suggesting an equilibrium of conformations between different states in solution, which could be one of the determinants affecting the binding kinetics of the ligand. Different ligands can undergo conformational selection or alter the equilibrium of conformations, inducing conformational rearrangements and resulting in different effects on binding kinetics. We then used molecular dynamics simulations to describe conformational changes of apo N-HSP90α in different conformational states. In summary, the study of the binding kinetics and molecular mechanisms of N-HSP90α provides valuable information for the development of more targeted therapeutic approaches.

Enhancing Multi-species Liver Microsomal Stability Prediction through Artificial Intelligence

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.

Machine Learning Prediction of Quantum Yields and Wavelengths of Aggregation-Induced Emission Molecules

The aggregation-induced emission (AIE) effect exhibits a significant influence on the development of luminescent materials and has made remarkable progress over the past decades. The advancement of high-performance AIE materials requires fast and accurate predictions of their photophysical properties, which is impeded by the inherent limitations of quantum chemical calculations. In this work, we present an accurate machine learning approach for the fast predictions of quantum yields and wavelengths to screen out AIE molecules. A database of about 563 organic luminescent molecules with quantum yields and wavelengths in the monomeric/aggregated states was established. Individual/combined molecular fingerprints were selected and compared elaborately to attain appropriate molecular descriptors. Different machine learning algorithms combined with favorable molecular fingerprints were further screened to achieve more accurate prediction models. The simulation results indicate that combined molecular fingerprints yield more accurate predictions in the aggregated states, and random forest and gradient boosting regression algorithms show the best predictions in quantum yields and wavelengths, respectively. Given the successful applications of machine learning in quantum yields and wavelengths, it is reasonable to anticipate that machine learning can serve as a complementary strategy to traditional experimental/theoretical methods in the investigation of aggregation-induced luminescent molecules to facilitate the discovery of luminescent materials.

A comparative study of the predictive performance of different descriptor calculation tools: Molecular-based elution order modeling and interpretation of retention mechanism for isomeric compounds from METLIN database

In the pharmaceutical industry, the need for analytical standards is a bottleneck for comprehensive evaluation and quality control of intermediate and end products. These are complex mixtures containing structurally related molecules. In this regard, chromatographic peak annotation, especially for critical pairs of isomers and closest structural analogs, can be supported by using a Quantitative Structure Retention Relationship (QSRR) approach. In our study, we investigated the fundamental basis of the reversed-phase (RP) retention mechanism for 1141 isomeric compounds from the METLIN SMRT dataset. Nine different descriptor calculation tools combined with different feature selection methods (genetic algorithm (GA), stepwise, Boruta) and machine learning (ML) approaches (support vector machine (SVM), multiple linear regression (MLR), random forest (RF), XGBoost) were applied to provide a reliable molecular structure-based interpretation of RP retention behaviour of the isomeric compounds. Strict internal and external validation metrics were used to select models with the best predictive capabilities (rtest > 0.73, order of elution > 60 %). For the developed models, mean absolute errors were in the range of 60 to 110 s. Stepwise and GA showed the most suitable performance as descriptor selection methods, while SVM and XGBoost modeling gave satisfactory predictive characteristics in most cases. Validation performed on the published experimental data for structurally related pharmaceutical compounds confirmed the best accuracy of MLR modeling in combination with GA feature selection of general physico-chemical properties. The resulting models will be useful for the prediction of separation and identification of structurally related compounds in pharmaceutical analysis, providing a simultaneous understanding of the interaction mechanisms leading to their retention under RP conditions.

Utilizing Machine Learning Models for Predicting Diamagnetic Susceptibility of Organic Compounds

This research is centered on examining the magnetic characteristics of organic molecules, with a particular emphasis on magnetic susceptibility, an essential physical property that provides insights into molecular microstructures and reaction processes. Traditional approaches for determining and calculating magnetic susceptibility are generally inefficient and demanding. To overcome these challenges, we have introduced a novel approach using quantitative structure-property relationships, which efficiently elucidates the relationship between the structural properties of molecules and their molar magnetic susceptibility. In our study, we utilized a comprehensive database comprising molar magnetic susceptibility data for 382 organic molecules. We applied six distinct molecular fingerprinting methods-RDKit Fingerprint, Morgan Fingerprint, MACCS Keys, atom pair fingerprint, Avalon Fingerprint, and topology fingerprint-as feature inputs for training seven different machine learning models, namely random forest, AdaBoost, gradient boosting, extra trees, elastic net, support vector machine, and multilayer perceptron (MLP). Our findings revealed that the integration of the atom pair fingerprint with the MLP model yielded R2 values of 0.88 and 0.90 in the validation and test sets, respectively, showcasing exceptional predictive accuracy. This advancement significantly expedites research and development processes related to the magnetic properties of organic molecules. Moreover, by employing this effective predictive method, it is expected to considerably reduce both experimental and computational expenses while maintaining high accuracy. This development represents a breakthrough in the rapid screening and prediction of properties for various compounds, offering a new and efficient pathway in this field of study.

Analysing the effect caused by increasing the molecular volume in M1-AChR receptor agonists and antagonists: a structural and computational study

M1 muscarinic acetylcholine receptor (M1-AChR), a member of the G protein-coupled receptors (GPCR) family, plays a crucial role in learning and memory, making it an important drug target for Alzheimer's disease (AD) and schizophrenia. M1-AChR activation and deactivation have shown modifying effects in AD and PD preclinical models, respectively. However, understanding the pharmacology associated with M1-AChR activation or deactivation is complex, because of the low selectivity among muscarinic subtypes, hampering their therapeutic applications. In this regard, we constructed two quantitative structure-activity relationship (QSAR) models, one for M1-AChR agonists (total and partial), and the other for the antagonists. The binding mode of 59 structurally different compounds, including agonists and antagonists with experimental binding affinity values (pKi), were analyzed employing computational molecular docking over different structures of M1-AChR. Furthermore, we considered the interaction energy (Einter), the number of rotatable bonds (NRB), and lipophilicity (ilogP) for the construction of the QSAR model for agonists (R2 = 89.64, QLMO2 = 78, and Qext2 = 79.1). For the QSAR model of antagonists (R2 = 88.44, QLMO2 = 82, and Qext2 = 78.1) we considered the Einter, the fraction of sp3 carbons fCsp3, and lipophilicity (MlogP). Our results suggest that the ligand volume is a determinant to establish its biological activity (agonist or antagonist), causing changes in binding energy, and determining the affinity for M1-AChR.

Natural Isatin Derivatives Against Black Fungus: In Silico Studies

During this coronavirus pandemic, when a lot of people are already severely afflicted with SARS-CoV-19, the dispersion of black fungus is making it worse, especially in the Indian subcontinent. Considering this situation, the idea for an in silico study to identify the potential inhibitor against black fungal infection is envisioned and computational analysis has been conducted with isatin derivatives that exhibit considerable antifungal activity. Through this in silico study, several pharmacokinetics properties like absorption, distribution, metabolism, excretion, and toxicity (ADMET) are estimated for various derivatives. Lipinski rules have been used to observe the drug likeliness property, and to study the electronic properties of the molecules, quantum mechanism was analyzed using the density functional theory (DFT). After applying molecular docking of the isatin derivatives with sterol 14-alpha demethylase enzyme of black fungus, a far higher docking affinity score has been observed for the isatin sulfonamide-34 (derivative 1) than the standard fluconazole. Lastly, molecular dynamic (MD) simulation has been performed for 100 ns to examine the stability of the proposed drug complex by estimating Root Mean Square Deviation (RMSD), Radius of gyration (Rg), Solvent accessible surface area (SASA), Root Mean Square Fluctuation (RMSF), as well as hydrogen bond. Listed ligands have precisely satisfied every pharmacokinetics requirement for a qualified drug candidate and they are non-toxic, non-carcinogenic, and have high stability. This natural molecule known as isatin derivative 1 has shown the potential of being a drug for fungal treatment. However, the impact of the chemicals on living cells requires more investigation and research.

Quinoline-based thiazolyl-hydrazones target cancer cells through autophagy inhibition

Heterocyclic pharmacophores such as thiazole and quinoline rings have a significant role in medicinal chemistry. They are considered privileged structures since they constitute several Food and Drug Administration (FDA)-approved drugs for cancer treatment. Herein, we report the synthesis, in silico evaluation of the ADMET profiles, and in vitro investigation of the anticancer activity of a series of novel thiazolyl-hydrazones based on the 8-quinoline (1a-c), 2-quinoline (2a-c), and 8-hydroxy-2-quinolyl moiety (3a-c). The panel of several human cancer cell lines and the nontumorigenic human embryonic kidney cell line HEK-293 were used to evaluate the compound-mediated in vitro anticancer activities, leading to [2-(2-(quinolyl-8-ol-2-ylmethylene)hydrazinyl)]-4-(4-methoxyphenyl)-1,3-thiazole (3c) as the most promising compound. The study revealed that 3c blocks the cell-cycle progression of a human colon cancer cell line (HCT-116) in the S phase and induces DNA double-strand breaks. Also, our findings demonstrate that 3c accumulates in lysosomes, ultimately leading to the cell death of the hepatocellular carcinoma cell line (Hep-G2) and HCT-116 cells, by the mechanism of autophagy inhibition.

In silico environmental risk assessment improves efficiency for pesticide safety management

Pesticides are indispensable to maintain crop quality and food production worldwide, but their use also poses environmental risks. Pesticide risk assessment involves a series of complex, expensive and time-consuming toxicity tests. To improve the efficiency and accuracy for assessing the environmental impact of pesticides, numerous computational tools have been developed. However, there is a notable deficiency in critical analysis or a systematic summary of environmental risk assessment tools and their applicable contexts. Here, many of the current approaches and tools for assessing environmental risks posed by pesticides are reviewed, and the question of whether these tools are fit for use on complex multicomponent scenarios is discussed. We analyze the adaptations of these tools to aquatic and terrestrial ecosystems, followed by the provision of resources for predicting pesticide concentrations in environmental medias, including air, soil and water. The successful application of computational tools for risk assessment and interpretation of predicted results will also be discussed. This assessment serves as a valuable resource, enabling scientists to utilize suitable models to enhance the robustness of pesticides risk assessments.

Advances in Drug Discovery and Design using Computer-aided Molecular Modeling

Computer-aided molecular modeling is a rapidly emerging technology that is being used to accelerate the discovery and design of new drug therapies. It involves the use of computer algorithms and 3D structures of molecules to predict interactions between molecules and their behavior in the body. This has drastically improved the speed and accuracy of drug discovery and design. Additionally, computer-aided molecular modeling has the potential to reduce costs, increase the quality of data, and identify promising targets for drug development. Through the use of sophisticated methods, such as virtual screening, molecular docking, pharmacophore modeling, and quantitative structure-activity relationships, scientists can achieve higher levels of efficacy and safety for new drugs. Moreover, it can be used to understand the activity of known drugs and simplify the process of formulating, optimizing, and predicting the pharmacokinetics of new and existing drugs. In conclusion, computer-aided molecular modeling is an effective tool to rapidly progress drug discovery and design by predicting the interactions between molecules and anticipating the behavior of new drugs in the body.

Predicting drug-induced liver injury using graph attention mechanism and molecular fingerprints

Drug-induced liver injury (DILI) is a significant issue in drug development and clinical treatment due to its potential to cause liver dysfunction or damage, which, in severe cases, can lead to liver failure or even fatality. DILI has numerous pathogenic factors, many of which remain incompletely understood. Consequently, it is imperative to devise methodologies and tools for anticipatory assessment of DILI risk in the initial phases of drug development. In this study, we present DMFPGA, a novel deep learning predictive model designed to predict DILI. To provide a comprehensive description of molecular properties, we employ a multi-head graph attention mechanism to extract features from the molecular graphs, representing characteristics at the level of compound nodes. Additionally, we combine multiple fingerprints of molecules to capture features at the molecular level of compounds. The fusion of molecular fingerprints and graph features can more fully express the properties of compounds. Subsequently, we employ a fully connected neural network to classify compounds as either DILI-positive or DILI-negative. To rigorously evaluate DMFPGA's performance, we conduct a 5-fold cross-validation experiment. The obtained results demonstrate the superiority of our method over four existing state-of-the-art computational approaches, exhibiting an average AUC of 0.935 and an average ACC of 0.934. We believe that DMFPGA is helpful for early-stage DILI prediction and assessment in drug development.

Mining Chemogenomic Spaces for Prediction of Drug-Target Interactions

The pipeline of drug discovery consists of a number of processes; drug-target interaction determination is one of the salient steps among them. Computational prediction of drug-target interactions can facilitate in reducing the search space of experimental wet lab-based verifications steps, thus considerably reducing time and other resources dedicated to the drug discovery pipeline. While machine learning-based methods are more widespread for drug-target interaction prediction, network-centric methods are also evolving. In this chapter, we focus on the process of the drug-target interaction prediction from the perspective of using machine learning algorithms and the various stages involved for developing an accurate predictor.

GUIDEMOL: A Python graphical user interface for molecular descriptors based on RDKit

GUIDEMOL is a Python computer program based on the RDKit software to process molecular structures and calculate molecular descriptors with a graphical user interface using the tkinter package. It can calculate descriptors already implemented in RDKit as well as grid representations of 3D molecular structures using the electrostatic potential or voxels. The GUIDEMOL app provides easy access to RDKit tools for chemoinformatics users with no programming skills and can be adapted to calculate other descriptors or to trigger other procedures. A command line interface (CLI) is also provided for the calculation of grid representations. The source code is available at https://github.com/jairesdesousa/guidemol.

Praziquantel Fifty Years on: A Comprehensive Overview of Its Solid State

This review discusses the entire progress made on the anthelmintic drug praziquantel, focusing on the solid state and, therefore, on anhydrous crystalline polymorphs, amorphous forms, and multicomponent systems (i.e., hydrates, solvates, and cocrystals). Despite having been extensively studied over the last 50 years, new polymorphs and the greater part of their cocrystals have only been identified in the past decade. Progress in crystal engineering science (e.g., the use of mechanochemistry as a solid form screening tool and more strategic structure-based methods), along with the development of analytical techniques, including Synchrotron X-ray analyses, spectroscopy, and microscopy, have furthered the identification of unknown crystal structures of the drug. Also, computational modeling has significantly contributed to the prediction and design of new cocrystals by considering structural conformations and interactions energy. Whilst the insights on praziquantel polymorphs discussed in the present review will give a significant contribution to controlling their formation during manufacturing and drug formulation, the detailed multicomponent forms will help in designing and implementing future praziquantel-based functional materials. The latter will hopefully overcome praziquantel's numerous drawbacks and exploit its potential in the field of neglected tropical diseases.

Navigating the Expansive Landscapes of Soft Materials: A User Guide for High-Throughput Workflows

Synthetic polymers are highly customizable with tailored structures and functionality, yet this versatility generates challenges in the design of advanced materials due to the size and complexity of the design space. Thus, exploration and optimization of polymer properties using combinatorial libraries has become increasingly common, which requires careful selection of synthetic strategies, characterization techniques, and rapid processing workflows to obtain fundamental principles from these large data sets. Herein, we provide guidelines for strategic design of macromolecule libraries and workflows to efficiently navigate these high-dimensional design spaces. We describe synthetic methods for multiple library sizes and structures as well as characterization methods to rapidly generate data sets, including tools that can be adapted from biological workflows. We further highlight relevant insights from statistics and machine learning to aid in data featurization, representation, and analysis. This Perspective acts as a "user guide" for researchers interested in leveraging high-throughput screening toward the design of multifunctional polymers and predictive modeling of structure-property relationships in soft materials.

TeM-DTBA: time-efficient drug target binding affinity prediction using multiple modalities with Lasso feature selection

Drug discovery, especially virtual screening and drug repositioning, can be accelerated through deeper understanding and prediction of Drug Target Interactions (DTIs). The advancement of deep learning as well as the time and financial costs associated with conventional wet-lab experiments have made computational methods for DTI prediction more popular. However, the majority of these computational methods handle the DTI problem as a binary classification task, ignoring the quantitative binding affinity that determines the drug efficacy to their target proteins. Moreover, computational space as well as execution time of the model is often ignored over accuracy. To address these challenges, we introduce a novel method, called Time-efficient Multimodal Drug Target Binding Affinity (TeM-DTBA), which predicts the binding affinity between drugs and targets by fusing different modalities based on compound structures and target sequences. We employ the Lasso feature selection method, which lowers the dimensionality of feature vectors and speeds up the proposed model training time by more than 50%. The results from two benchmark datasets demonstrate that our method outperforms state-of-the-art methods in terms of performance. The mean squared errors of 18.8% and 23.19%, achieved on the KIBA and Davis datasets, respectively, suggest that our method is more accurate in predicting drug-target binding affinity.

From intuition to AI: evolution of small molecule representations in drug discovery

Within drug discovery, the goal of AI scientists and cheminformaticians is to help identify molecular starting points that will develop into safe and efficacious drugs while reducing costs, time and failure rates. To achieve this goal, it is crucial to represent molecules in a digital format that makes them machine-readable and facilitates the accurate prediction of properties that drive decision-making. Over the years, molecular representations have evolved from intuitive and human-readable formats to bespoke numerical descriptors and fingerprints, and now to learned representations that capture patterns and salient features across vast chemical spaces. Among these, sequence-based and graph-based representations of small molecules have become highly popular. However, each approach has strengths and weaknesses across dimensions such as generality, computational cost, inversibility for generative applications and interpretability, which can be critical in informing practitioners' decisions. As the drug discovery landscape evolves, opportunities for innovation continue to emerge. These include the creation of molecular representations for high-value, low-data regimes, the distillation of broader biological and chemical knowledge into novel learned representations and the modeling of up-and-coming therapeutic modalities.