QSAR modeling: where have you been? Where are you going to?

Estrogen Receptor Alpha Binders for Hormone-Dependent Forms of Breast Cancer: e-QSAR and Molecular Docking Supported by X-ray Resolved Structures

Cancer, a life-disturbing and lethal disease with a high global impact, causes significant economic, social, and health challenges. Breast cancer refers to the abnormal growth of cells originating from breast tissues. Hormone-dependent forms of breast cancer, such as those influenced by estrogen, prompt the exploration of estrogen receptors as targets for potential therapeutic interventions. In this study, we conducted e-QSAR molecular docking and molecular dynamics analyses on a diverse set of inhibitors targeting estrogen receptor alpha (ER-α). The e-QSAR model is based on a genetic algorithm combined with multilinear regression analysis. The newly developed model possesses a balance between predictive accuracy and mechanistic insights adhering to the OECD guidelines. The e-QSAR model pointed out that sp2-hybridized carbon and nitrogen atoms are important atoms governing binding profiles. In addition, a specific combination of H-bond donors and acceptors with carbon, nitrogen, and ring sulfur atoms also plays a crucial role. The results are supported by molecular docking, MD simulations, and X-ray-resolved structures. The novel results could be useful for future drug development for ER-α.

Gene Target Prediction of Environmental Chemicals Using Coupled Matrix-Matrix Completion

Human exposure to toxic chemicals presents a huge health burden. Key to understanding chemical toxicity is knowledge of the molecular target(s) of the chemicals. Because a comprehensive safety assessment for all chemicals is infeasible due to limited resources, a robust computational method for discovering targets of environmental exposures is a promising direction for public health research. In this study, we implemented a novel matrix completion algorithm named coupled matrix-matrix completion (CMMC) for predicting direct and indirect exposome-target interactions, which exploits the vast amount of accumulated data regarding chemical exposures and their molecular targets. Our approach achieved an AUC of 0.89 on a benchmark data set generated using data from the Comparative Toxicogenomics Database. Our case studies with bisphenol A and its analogues, PFAS, dioxins, PCBs, and VOCs show that CMMC can be used to accurately predict molecular targets of novel chemicals without any prior bioactivity knowledge. Our results demonstrate the feasibility and promise of computationally predicting environmental chemical-target interactions to efficiently prioritize chemicals in hazard identification and risk assessment.

Recent advances in the area of plant-based anti-cancer drug discovery using computational approaches

Phytocompounds are a well-established source of drug discovery due to their unique chemical and functional diversities. In the area of cancer therapeutics, several phytocompounds have been used till date to design and develop new drugs. One of the desired interests of pharmaceutical companies and researchers globally is that new anti-cancer leads are discovered, for which phytocompounds can be considered a valuable source. Simultaneously, in recent years, the growth of computational approaches like virtual screening (VS), molecular dynamics (MD), pharmacophore modelling, Quantitative structure-activity relationship (QSAR), Absorption Distribution Metabolism Excretion and Toxicity (ADMET), network biology, and machine learning (ML) has gained importance due to their efficiency, reduced time-consuming nature, and cost-effectiveness. Therefore, the present review amalgamates the information on plant-based molecules identified for cancer lead discovery from in silico approaches. The mandate of this review is to discuss studies published in the last 5-6 years that aim to identify the phytomolecules as leads against cancer with the help of traditional computational approaches as well as newer techniques like network pharmacology and ML. This review also lists the databases and webservers available in the public domain for phytocompounds related information that can be harnessed for drug discovery. It is expected that the present review would be useful to pharmacologists, medicinal chemists, molecular biologists, and other researchers involved in the development of natural products (NPs) into clinically effective lead molecules.

Transformers enable accurate prediction of acute and chronic chemical toxicity in aquatic organisms

Environmental hazard assessments are reliant on toxicity data that cover multiple organism groups. Generating experimental toxicity data is, however, resource-intensive and time-consuming. Computational methods are fast and cost-efficient alternatives, but the low accuracy and narrow applicability domains have made their adaptation slow. Here, we present a AI-based model for predicting chemical toxicity. The model uses transformers to capture toxicity-specific features directly from the chemical structures and deep neural networks to predict effect concentrations. The model showed high predictive performance for all tested organism groups-algae, aquatic invertebrates and fish-and has, in comparison to commonly used QSAR methods, a larger applicability domain and a considerably lower error. When the model was trained on data with multiple effect concentrations (EC50/EC10), the performance was further improved. We conclude that deep learning and transformers have the potential to markedly advance computational prediction of chemical toxicity.

Relationships between the Photodegradation Reaction Rate and Structural Properties of Polymer Systems

The development of reusable polymeric materials inspires an attempt to combine renewable biomass with upcycling to form a biorenewable closed system. It has been reported that 2,5-furandicarboxylic acid (FDCA) can be recovered for recycling when incorporated as monomers into photodegradable polymeric systems. Here, we develop a procedure to better understand the photodegradation reactions combining density functional theory (DFT) based time-dependent excited-state molecular dynamics (TDESMD) studies with machine learning-based quantitative structure-activity relationships (QSAR) methodology. This procedure allows for the unveiling of hidden structural features between active orbitals that affect the rate of photodegradation and is coined InfoTDESMD. Findings show that electrotopological features are influential factors affecting the rate of photodegradation in differing environments. Additionally, statistical validations and knowledge-based analysis of descriptors are conducted to further understand the structural features' influence on the rate of photodegradation of polymeric materials.

AI-enhanced chemical paradigm: From molecular graphs to accurate prediction and mechanism

The development of accurate and interpretable models for predicting reaction constants of organic compounds with hydroxyl radicals is vital for advancing quantitative structure-activity relationships (QSAR) in pollutant degradation. Methods like molecular descriptors, molecular fingerprinting, and group contribution methods have limitations, as traditional machine learning struggles to capture all intramolecular information simultaneously. To address this, we established an integrated graph neural network (GNN) with approximately 12 million learnable parameters. GNN represents atoms as nodes and chemical bonds as edges, thus transforming molecules into a graph structures, effectively capturing microscopic properties while depicting atom connectivity in non-Euclidean space. Our datasets comprise 1401 pollutants to develop an integrated GNN model with Bayesian optimization, the model achieves root mean square errors of 0.165, 0.172, and 0.189 on the training, validation, and test datasets, respectively. Furthermore, we assess molecular structure similarity using molecular fingerprint to enhance the model's applicability. Afterwards, we propose a gradient weight mapping method for model explainability, uncovering the key functional groups in chemical reactions in artificial intelligence perspective, which would boost chemistry through artificial intelligence extreme arithmetic power.

Monte Carlo Optimization Method Based QSAR Modeling of Placental Barrier Permeability

PURPOSE

In order to ensure that drug administration is safe during pregnancy, it is crucial to have the possibility to predict the placental permeability of drugs in humans. The experimental method which is most widely used for the said purpose is in vitro human placental perfusion, though the approach is highly expensive and time consuming. Quantitative structure-activity relationship (QSAR) modeling represents a powerful tool for the assessment of the drug placental transfer, and can be successfully employed to be an alternative in in vitro experiments.

METHODS

The conformation-independent QSAR models covered in the present study were developed through the use of the SMILES notation descriptors and local molecular graph invariants. What is more, the Monte Carlo optimization method, was used in the test sets and the training sets as the model developer with three independent molecular splits.

RESULTS

A range of different statistical parameters was used to validate the developed QSAR model, including the standard error of estimation, mean absolute error, root-mean-square error (RMSE), correlation coefficient, cross-validated correlation coefficient, Fisher ratio, MAE-based metrics and the correlation ideality index. Once the mentioned statistical methods were employed, an excellent predictive potential and robustness of the developed QSAR model was demonstrated. In addition, the molecular fragments, which are derived from the SMILES notation descriptors accounting for the decrease or increase in the investigated activity, were revealed.

CONCLUSION

The presented QSAR modeling can be an invaluable tool for the high-throughput screening of the placental permeability of drugs.

Development of a hierarchical support vector regression-based in silico model for the prediction of the cysteine depletion in DPRA

Topical and transdermal treatments have been dramatically growing recently and it is crucial to consider skin sensitization during the drug discovery and development process for these administration routes. Various tests, including animal and non-animal approaches, have been devised to assess the potential for skin sensitization. Furthermore, numerous in silico models have been created, providing swift and cost-effective alternatives to traditional methods such as in vivo, in vitro, and in chemico methods for categorizing compounds. In this study, a quantitative structure-activity relationship (QSAR) model was developed using the innovative hierarchical support vector regression (HSVR) scheme. The aim was to quantitatively predict the potential for skin sensitization by analyzing the percent of cysteine depletion in Direct Peptide Reactivity Assay (DPRA). The results demonstrated accurate, consistent, and robust predictions in the training set, test set, and outlier set. Consequently, this model can be employed to estimate skin sensitization potential of novel or virtual compounds.

QSAR modeling approaches to identify a novel ACE2 inhibitor that selectively bind with the C and N terminals of the ectodomain

Due to the high rates of drug development failure and the massive expenses associated with drug discovery, repurposing existing drugs has become more popular. As a result, we have used QSAR modelling on a large and varied dataset of 657 compounds in an effort to discover both explicit and subtle structural features requisite for ACE2 inhibitory activity, with the goal of identifying novel hit molecules. The QSAR modelling yielded a statistically robust QSAR model with high predictivity (R2tr=0.84, R2ex=0.79), previously undisclosed features, and novel mechanistic interpretations. The developed QSAR model predicted the ACE2 inhibitory activity (PIC50) of 1615 ZINC FDA compounds. This led to the detection of a PIC50 of 8.604 M for the hit molecule (ZINC000027990463). The hit molecule's docking score is -9.67 kcal/mol (RMSD 1.4). The hit molecule revealed 25 interactions with the residue ASP40, which defines the N and C termini of the ectodomain of ACE2. The HIT molecule conducted more than thirty contacts with water molecules and exhibited polar interaction with the ARG522 residue coupled with the second chloride ion, which is 10.4 nm away from the zinc ion. Both molecular docking and QSAR produced comparable findings. Moreover, MD simulation and MMGBSA studies verified docking analysis. The MD simulation showed that the hit molecule-ACE2 receptor complex is stable for 400 ns, suggesting that repurposed hit molecule 3 is a viable ACE2 inhibitor.

Beyond Amyloid: A Machine Learning-Driven Approach Reveals Properties of Potent GSK-3β Inhibitors Targeting Neurofibrillary Tangles

Current treatments for Alzheimer's disease (AD) focus on slowing memory and cognitive decline, but none offer curative outcomes. This study aims to explore and curate the common properties of active, drug-like molecules that modulate glycogen synthase kinase 3β (GSK-3β), a well-documented kinase with increased activity in tau hyperphosphorylation and neurofibrillary tangles-hallmarks of AD pathology. Leveraging quantitative structure-activity relationship (QSAR) data from the PubChem and ChEMBL databases, we employed seven machine learning models: logistic regression (LogR), k-nearest neighbors (KNN), random forest (RF), support vector machine (SVM), extreme gradient boosting (XGB), neural networks (NNs), and ensemble majority voting. Our goal was to correctly predict active and inactive compounds that inhibit GSK-3β activity and identify their key properties. Among the six individual models, the NN demonstrated the highest performance with a 79% AUC-ROC on unbalanced external validation data, while the SVM model was superior in accurately classifying the compounds. The SVM and RF models surpassed NN in terms of Kappa values, and the ensemble majority voting model demonstrated slightly better accuracy to the NN on the external validation data. Feature importance analysis revealed that hydrogen bonds, phenol groups, and specific electronic characteristics are important features of molecular descriptors that positively correlate with active GSK-3β inhibition. Conversely, structural features like imidazole rings, sulfides, and methoxy groups showed a negative correlation. Our study highlights the significance of structural, electronic, and physicochemical descriptors in screening active candidates against GSK-3β. These predictive features could prove useful in therapeutic strategies to understand the important properties of GSK-3β candidate inhibitors that may potentially benefit non-amyloid-based AD treatments targeting neurofibrillary tangles.

Mechanistic QSAR modeling derived virtual screening, drug repurposing, ADMET and in-vitro evaluation to identify anticancer lead as lysine-specific demethylase 5a inhibitor

A lysine-specific demethylase is an enzyme that selectively eliminates methyl groups from lysine residues. KDM5A, also known as JARID1A or RBP2, belongs to the KDM5 Jumonji histone demethylase subfamily. To identify novel molecules that interact with the LSD5A receptor, we created a quantitative structure-activity relationship (QSAR) model. A group of 435 compounds was used in a study of the quantitative relationship between structure and activity to guess the IC50 values for blocking LASD5A. We used a genetic algorithm-multilinear regression-based quantitative structure-activity connection model to forecast the bioactivity (PIC50) of 1615 food and drug administration pharmaceuticals from the zinc database with the goal of repurposing clinically used medications. We used molecular docking, molecular dynamic simulation modelling, and molecular mechanics generalised surface area analysis to investigate the molecule's binding mechanism. A genetic algorithm and multi-linear regression method were used to make six variable-based quantitative structure-activity relationship models that worked well (R2 = 0.8521, Q2LOO = 0.8438, and Q2LMO = 0.8414). ZINC000000538621 was found to be a new hit against LSD5A after a quantitative structure-activity relationship-based virtual screening of 1615 zinc food and drug administration compounds. The docking analysis revealed that the hit molecule 11 in the KDM5A binding pocket adopted a conformation similar to the pdb-6bh1 ligand (docking score: -8.61 kcal/mol). The results from molecular docking and the quantitative structure-activity relationship were complementary and consistent. The most active lead molecule 11, which has shown encouraging results, has good absorption, distribution, metabolism, and excretion (ADME) properties, and its toxicity has been shown to be minimal. In addition, the MTT assay of ZINC000000538621 with MCF-7 cell lines backs up the in silico studies. We used molecular mechanics generalise borne surface area analysis and a 200-ns molecular dynamics simulation to find structural motifs for KDM5A enzyme interactions. Thus, our strategy will likely expand food and drug administration molecule repurposing research to find better anticancer drugs and therapies.Communicated by Ramaswamy H. Sarma.

A Primer on 2D Descriptors in Selectivity Modeling for Asymmetric Catalysis

Machine learning has permeated all fields of research, including chemistry, and is now an integral part of the design of novel compounds with desired properties. In the field of asymmetric catalysis, the preference still lies with models based on a physical understanding of the catalysis phenomenon and the electronic and steric properties of catalysts. However, such models require quantum chemical calculations and are thus limited by their computational cost. Here, we highlight the recent advances in modeling catalyst selectivity by using the 2D structures of catalysts and substrates. While these have a less explicit mechanistic connection to the modeled property, 2D descriptors, such as topological indices, molecular fingerprints, and fragments, offer the tremendous advantages of low cost and high speed of calculations. This makes them optimal for the in-silico screening of large amounts of data. We provide an overview of common quantitative structure-property relationship workflow, model building and validation techniques, applications of these methodologies in asymmetric catalysis design, and an outlook on improving the understanding of 2D-based models.

Harmonization Risks and Rewards: Nano-QSAR for Agricultural Nanomaterials

This comprehensive review explores the emerging landscape of Nano-QSAR (quantitative structure-activity relationship) for assessing the risk and potency of nanomaterials in agricultural settings. The paper begins with an introduction to Nano-QSAR, providing background and rationale, and explicitly states the hypotheses guiding the review. The study navigates through various dimensions of nanomaterial applications in agriculture, encompassing their diverse properties, types, and associated challenges. Delving into the principles of QSAR in nanotoxicology, this article elucidates its application in evaluating the safety of nanomaterials, while addressing the unique limitations posed by these materials. The narrative then transitions to the progression of Nano-QSAR in the context of agricultural nanomaterials, exemplified by insightful case studies that highlight both the strengths and the limitations inherent in this methodology. Emerging prospects and hurdles tied to Nano-QSAR in agriculture are rigorously examined, casting light on important pathways forward, existing constraints, and avenues for research enhancement. Culminating in a synthesis of key insights, the review underscores the significance of Nano-QSAR in shaping the future of nanoenabled agriculture. It provides strategic guidance to steer forthcoming research endeavors in this dynamic field.

Rethinking the applicability domain analysis in QSAR models

Notwithstanding the wide adoption of the OECD principles (or best practices) for QSAR modeling, disparities between in silico predictions and experimental results are frequent, suggesting that model predictions are often too optimistic. Of these OECD principles, the applicability domain (AD) estimation has been recognized in several reports in the literature to be one of the most challenging, implying that the actual reliability measures of model predictions are often unreliable. Applying tree-based error analysis workflows on 5 QSAR models reported in the literature and available in the QsarDB repository, i.e., androgen receptor bioactivity (agonists, antagonists, and binders, respectively) and membrane permeability (highest membrane permeability and the intrinsic permeability), we demonstrate that predictions erroneously tagged as reliable (AD prediction errors) overwhelmingly correspond to instances in subspaces (cohorts) with the highest prediction error rates, highlighting the inhomogeneity of the AD space. In this sense, we call for more stringent AD analysis guidelines which require the incorporation of model error analysis schemes, to provide critical insight on the reliability of underlying AD algorithms. Additionally, any selected AD method should be rigorously validated to demonstrate its suitability for the model space over which it is applied. These steps will ultimately contribute to more accurate estimations of the reliability of model predictions. Finally, error analysis may also be useful in "rational" model refinement in that data expansion efforts and model retraining are focused on cohorts with the highest error rates.

The pursuit of accurate predictive models of the bioactivity of small molecules

Property prediction is a key interest in chemistry. For several decades there has been a continued and incremental development of mathematical models to predict properties. As more data is generated and accumulated, there seems to be more areas of opportunity to develop models with increased accuracy. The same is true if one considers the large developments in machine and deep learning models. However, along with the same areas of opportunity and development, issues and challenges remain and, with more data, new challenges emerge such as the quality and quantity and reliability of the data, and model reproducibility. Herein, we discuss the status of the accuracy of predictive models and present the authors' perspective of the direction of the field, emphasizing on good practices. We focus on predictive models of bioactive properties of small molecules relevant for drug discovery, agrochemical, food chemistry, natural product research, and related fields.

Creating New Antifoulants Using the Tools and Tactics of Medicinal Chemistry

The unwanted accumulation of marine micro- and macroorganisms such as algae and barnacles on submerged man-made structures and vessel hulls is a major challenge for any marine operation. Known as biofouling, this problem leads to reduced hydrodynamic efficiency, significantly increased fuel usage, microbially induced corrosion, and, if not managed appropriately, eventual loss of both performance and structural integrity. Ship hull biofouling in the international maritime transport network conservatively accounts for 0.6% of global carbon emissions, highlighting the global scale and the importance of this problem. Improved antifouling strategies to limit surface colonization are paramount for essential activities such as shipping, aquaculture, desalination, and the marine renewable energy sector, representing both a multibillion dollar cost and a substantial practical challenge. From an ecological perspective, biofouling is a primary contributor to the global spread of invasive marine species, which has extensive implications for the marine environment.Historically, heavy metal-based toxic biocides have been used to control biofouling. However, their unwanted collateral ecological damage on nontarget species and bioaccumulation has led to recent global bans. With expanding human activities within aquaculture and offshore energy, it is both urgent and apparent that environmentally friendly surface protection remains key for maintaining the function of both moving and stationary marine structures. Biofouling communities are typically a highly complex network of both micro- and macroorganisms, representing a broad section of life from bacteria to macrophytes and animals. Given this diversity, it is unrealistic to expect that a single antifouling "silver bullet" will prevent colonization with the exception of generally toxic biocides. For that reason, modern and future antifouling solutions are anticipated to rely on novel coating technologies and "combination therapies" where mixtures of narrow-spectrum bioactive components are used to provide coverage across fouling species. In contrast to the existing cohort of outdated, toxic antifouling strategies, such as copper- and tributyltin-releasing paints, modern drug discovery techniques are increasingly being employed for the rational design of effective yet safe alternatives. The challenge for a medicinal chemistry approach is to effectively account for the large taxonomic diversity among fouling organisms combined with a lack of well-defined conserved molecular targets within most taxa.The current Account summarizes our work employing the tools of modern medicinal chemistry to discover, modify, and develop optimized and scalable antifouling solutions based on naturally occurring antifouling and repelling compounds from both marine and terrestrial sources. Inspiration for rational design comes from targeted studies on allelopathic natural products, natural repelling peptides, and secondary metabolites from sessile marine organisms with clean exteriors, which has yielded several efficient and promising antifouling leads.

Integrating QSAR modelling and deep learning in drug discovery: the emergence of deep QSAR

Quantitative structure-activity relationship (QSAR) modelling, an approach that was introduced 60 years ago, is widely used in computer-aided drug design. In recent years, progress in artificial intelligence techniques, such as deep learning, the rapid growth of databases of molecules for virtual screening and dramatic improvements in computational power have supported the emergence of a new field of QSAR applications that we term 'deep QSAR'. Marking a decade from the pioneering applications of deep QSAR to tasks involved in small-molecule drug discovery, we herein describe key advances in the field, including deep generative and reinforcement learning approaches in molecular design, deep learning models for synthetic planning and the application of deep QSAR models in structure-based virtual screening. We also reflect on the emergence of quantum computing, which promises to further accelerate deep QSAR applications and the need for open-source and democratized resources to support computer-aided drug design.

The role and potential of computer-aided drug discovery strategies in the discovery of novel antimicrobials

Antimicrobial resistance (AMR) has become more of a concern in recent decades, particularly in infections associated with global public health threats. The development of new antibiotics is crucial to ensuring infection control and eradicating AMR. Although drug discovery and development are essential processes in the transformation of a drug candidate from the laboratory to the bedside, they are often very complicated, expensive, and time-consuming. The pharmaceutical sector is continuously innovating strategies to reduce research costs and accelerate the development of new drug candidates. Computer-aided drug discovery (CADD) has emerged as a powerful and promising technology that renews the hope of researchers for the faster identification, design, and development of cheaper, less resource-intensive, and more efficient drug candidates. In this review, we discuss an overview of AMR, the potential, and limitations of CADD in AMR drug discovery, and case studies of the successful application of this technique in the rapid identification of various drug candidates. This review will aid in achieving a better understanding of available CADD techniques in the discovery of novel drug candidates against resistant pathogens and other infectious agents.

Clustering of atoms relative to vector space in the Z-matrix coordinate system and 'graphical fingerprint' analysis of 3D pharmacophore structure

The behavior of a molecule within its environment is governed by chemical fields present in 3D space. However, beyond local descriptors in 3D, the conformations a molecule assumes, and the resulting clusters also play a role in influencing structure-activity models. This study focuses on the clustering of atoms according to the vector space of four atoms aligned in the Z-Matrix Reference system for molecular similarity. Using 3D-QSAR analysis, it was aimed to determine the pharmacophore groups as interaction points in the binding region of the β2-adrenoceptor target of fenoterol stereoisomers. Different types of local reactive descriptors of ligands have been used to elucidate points of interaction with the target. Activity values for ligand-receptor interaction energy were determined using the Levenberg-Marquardt algorithm. Using the Molecular Comparative Electron Topology method, the 3D pharmacophore model (3D-PhaM) was obtained after aligning and superimposing the molecules and was further validated by the molecular docking method. Best guesses were calculated with a non-output validation (LOO-CV) method. Finally, the data were calculated using the 'graphic fingerprint' technique. Based on the eLKlopman (Electrostatic LUMO Klopman) descriptor, the Q2 value of this derivative set was calculated as 0.981 and the R2ext value is calculated as 0.998.

Reliable method for predicting the binding affinity of RNA-small molecule interactions using machine learning

Ribonucleic acids (RNAs) play important roles in cellular regulation. Consequently, dysregulation of both coding and non-coding RNAs has been implicated in several disease conditions in the human body. In this regard, a growing interest has been observed to probe into the potential of RNAs to act as drug targets in disease conditions. To accelerate this search for disease-associated novel RNA targets and their small molecular inhibitors, machine learning models for binding affinity prediction were developed specific to six RNA subtypes namely, aptamers, miRNAs, repeats, ribosomal RNAs, riboswitches and viral RNAs. We found that differences in RNA sequence composition, flexibility and polar nature of RNA-binding ligands are important for predicting the binding affinity. Our method showed an average Pearson correlation (r) of 0.83 and a mean absolute error of 0.66 upon evaluation using the jack-knife test, indicating their reliability despite the low amount of data available for several RNA subtypes. Further, the models were validated with external blind test datasets, which outperform other existing quantitative structure-activity relationship (QSAR) models. We have developed a web server to host the models, RNA-Small molecule binding Affinity Predictor, which is freely available at: https://web.iitm.ac.in/bioinfo2/RSAPred/.

Developing a Novel Read-Across Concept for Ecotoxicological Risk Assessment of Phosphate Chemicals: A Case Study

This study introduces a novel concept approach for a read-across assessment, considering species sensitivity differences among phosphate chemicals within structurally similar compound groups. Twenty-five organic chemicals, with a log Kow of 5 or less, were categorized into three functional groups based on acetylcholinesterase (AChE) inhibition as a specific mode of action (MOA). The short-term aquatic toxicity data (LC50) for fish, crustaceans, and insects were collected from the U.S. EPA Ecotoxicology (ECOTOX) Knowledgebase. A geometric mean calculation method was applied for multiple toxic endpoints. Performance metrics for the new read-across concept, including correlation coefficient, bias, precision, and accuracy, were calculated. Overall, a slightly higher overestimation (49.2%) than underestimation (48.4%) in toxicity predictions was observed in two case studies. In Case study I, a strong positive correlation (r = 0.93) between the predicted and known toxicity values of target chemicals was observed, while in Case study II, with limited information on species and their ecotoxicity, showed a moderate correlation (r = 0.75). Overall, the bias and precision for Case study I were 0.32 ± 0.01, while Case study II showed 0.65 ± 0.06; however, the relative bias (%) increased from 37.65% (Case study I) to 91.94% (Case study II). Bland-Altman plots highlight the mean differences of 1.33 (Case study I) and 1.24 (Case study II), respectively. The new read-across concept, focusing on AChE inhibition and structural similarity, demonstrated good reliability, applicability, and accuracy with minimal bias. Future studies are needed to evaluate various types of chemical substances, diverse modes of action, functional groups, toxic endpoints, and test species to ensure overall comprehensiveness and robustness in toxicity predictions.

Artificial Intelligence and Machine Learning Methods to Evaluate Cardiotoxicity following the Adverse Outcome Pathway Frameworks

Cardiovascular disease is a leading global cause of mortality. The potential cardiotoxic effects of chemicals from different classes, such as environmental contaminants, pesticides, and drugs can significantly contribute to effects on health. The same chemical can induce cardiotoxicity in different ways, following various Adverse Outcome Pathways (AOPs). In addition, the potential synergistic effects between chemicals further complicate the issue. In silico methods have become essential for tackling the problem from different perspectives, reducing the need for traditional in vivo testing, and saving valuable resources in terms of time and money. Artificial intelligence (AI) and machine learning (ML) are among today's advanced approaches for evaluating chemical hazards. They can serve, for instance, as a first-tier component of Integrated Approaches to Testing and Assessment (IATA). This study employed ML and AI to assess interactions between chemicals and specific biological targets within the AOP networks for cardiotoxicity, starting with molecular initiating events (MIEs) and progressing through key events (KEs). We explored methods to encode chemical information in a suitable way for ML and AI. We started with commonly used approaches in Quantitative Structure-Activity Relationship (QSAR) methods, such as molecular descriptors and different types of fingerprint. We then increased the complexity of encoders, incorporating graph-based methods, auto-encoders, and character embeddings employed in neural language processing. We also developed a multimodal neural network architecture, capable of considering the complementary nature of different chemical representations simultaneously. The potential of this approach, compared to more conventional architectures designed to handle a single encoder, becomes apparent when the amount of data increases.

Anticancer Drug Discovery Based on Natural Products: From Computational Approaches to Clinical Studies

Globally, malignancies cause one out of six mortalities, which is a serious health problem. Cancer therapy has always been challenging, apart from major advances in immunotherapies, stem cell transplantation, targeted therapies, hormonal therapies, precision medicine, and palliative care, and traditional therapies such as surgery, radiation therapy, and chemotherapy. Natural products are integral to the development of innovative anticancer drugs in cancer research, offering the scientific community the possibility of exploring novel natural compounds against cancers. The role of natural products like Vincristine and Vinblastine has been thoroughly implicated in the management of leukemia and Hodgkin's disease. The computational method is the initial key approach in drug discovery, among various approaches. This review investigates the synergy between natural products and computational techniques, and highlights their significance in the drug discovery process. The transition from computational to experimental validation has been highlighted through in vitro and in vivo studies, with examples such as betulinic acid and withaferin A. The path toward therapeutic applications have been demonstrated through clinical studies of compounds such as silvestrol and artemisinin, from preclinical investigations to clinical trials. This article also addresses the challenges and limitations in the development of natural products as potential anti-cancer drugs. Moreover, the integration of deep learning and artificial intelligence with traditional computational drug discovery methods may be useful for enhancing the anticancer potential of natural products.

Unraveling the Push-Pull Effect in Acenes, Polyenes and Polyynes

Substituent effects (SEs) are fundamental for predicting molecular reactivity, while polyene, polyyne and acene derivatives are precursors to compounds with diverse applications. Computations were performed for Y-R-X systems, where reaction sites Y=NO2 and O- , substituents X=NO2 , CN, Cl, H, OH, NH2 , and spacers R=polyene, polyyne (n=1-5, 10 repeating units) and acene (up to tetracene). The cSAR (charge of the substituent active region) approach allowed to present, for the first time, quantitative relations describing the spacer's electron-donating and withdrawing properties as a function of n and the spacer type. The electronic properties of the X substituents depend on the type of spacer, its length and the Y group, which is an example of the reverse SE. To describe how the SE between Y and X weakens with n, two approaches were compared: cSAR and SESE (SE stabilization energy). The EDDB (electron density of delocalized bonds) characterize changes in electron delocalization in spacers due to the SE. A new approach - EDDB differential maps - allow to extract the effect of X substitution on the electron delocalization. The charges at spacer's C atoms correlate with cSAR; changes in the slopes confirm the charge transfer by resonance.

Bioactivity predictions and virtual screening using machine learning predictive model

Recently, there has been significant attention on machine learning algorithms for predictive modeling. Prediction models for enzyme inhibitors are limited, and it is essential to account for chemical biases while developing them. The lack of repeatability in available models and chemical bias issues constrain drug discovery and development. A new prediction model for enzyme inhibitors has been developed, and the model efficacy was checked using Dipeptidyl peptidase 4 (DPP-4) inhibitors. A Python script was prepared and can be provided for personal use upon request. Among various machine learning algorithms, it was found that Random Forest offers the best accuracy. Two models were compared, one with diverse training and test data and the other with a random split. It was concluded that machine learning predictive models based on the Murcko scaffold can address chemical bias concerns. In-silico screening of the Drug Bank database identified two molecules against DPP-4, which are previously proven hit molecules. The approach was further validated through molecular docking studies and molecular dynamics simulations, demonstrating the credibility and relevance of the developed model for future investigations and potential translation into clinical applications.Communicated by Ramaswamy H. Sarma.

Viscosity of Ionic Liquids: Theories and Models

Ionic liquids (ILs) offer a wide range of promising applications due to their unique and designable properties compared to conventional solvents. Further development and application of ILs require correlating/predicting their pressure-viscosity-temperature behavior. In this review, we firstly introduce methods for calculation of thermodynamic inputs of viscosity models. Next, we introduce theories, theoretical and semi-empirical models coupling various theories with EoSs or activity coefficient models, and empirical and phenomenological models for viscosity of pure ILs and IL-related mixtures. Our modelling description is followed immediately by model application and performance. Then, we propose simple predictive equations for viscosity of IL-related mixtures and systematically compare performances of the above-mentioned theories and models. In concluding remarks, we recommend robust predictive models for viscosity at atmospheric pressure as well as proper and consistent theories and models for P-η-T behavior. The work that still remains to be done to obtain the desired theories and models for viscosity of ILs and IL-related mixtures is also presented. The present review is structured from pure ILs to IL-related mixtures and aims to summarize and quantitatively discuss the recent advances in theoretical and empirical modelling of viscosity of ILs and IL-related mixtures.

Computational Nanotoxicology Models for Environmental Risk Assessment of Engineered Nanomaterials

Although engineered nanomaterials (ENMs) have tremendous potential to generate technological benefits in numerous sectors, uncertainty on the risks of ENMs for human health and the environment may impede the advancement of novel materials. Traditionally, the risks of ENMs can be evaluated by experimental methods such as environmental field monitoring and animal-based toxicity testing. However, it is time-consuming, expensive, and impractical to evaluate the risk of the increasingly large number of ENMs with the experimental methods. On the contrary, with the advancement of artificial intelligence and machine learning, in silico methods have recently received more attention in the risk assessment of ENMs. This review discusses the key progress of computational nanotoxicology models for assessing the risks of ENMs, including material flow analysis models, multimedia environmental models, physiologically based toxicokinetics models, quantitative nanostructure-activity relationships, and meta-analysis. Several challenges are identified and a perspective is provided regarding how the challenges can be addressed.

Curated mode-of-action data and effect concentrations for chemicals relevant for the aquatic environment

Chemicals in the aquatic environment can be harmful to organisms and ecosystems. Knowledge on effect concentrations as well as on mechanisms and modes of interaction with biological molecules and signaling pathways is necessary to perform chemical risk assessment and identify toxic compounds. To this end, we developed criteria and a pipeline for harvesting and summarizing effect concentrations from the US ECOTOX database for the three aquatic species groups algae, crustaceans, and fish and researched the modes of action of more than 3,300 environmentally relevant chemicals in literature and databases. We provide a curated dataset ready to be used for risk assessment based on monitoring data and the first comprehensive collection and categorization of modes of action of environmental chemicals. Authorities, regulators, and scientists can use this data for the grouping of chemicals, the establishment of meaningful assessment groups, and the development of in vitro and in silico approaches for chemical testing and assessment.

Machine Learning-Assisted Computational Screening of Adhesive Molecules Derived from Dihydroxyphenyl Alanine

Marine mussels adhere to virtually any surface via 3,4-dihydroxyphenyl-L-alanines (L-DOPA), an amino acid largely contained in their foot proteins. The biofriendly, water-repellent, and strong adhesion of L-DOPA are unparalleled by any synthetic adhesive. Inspired by this, we computationally designed diverse derivatives of DOPA and studied their potential as adhesives or coating materials. We used first-principles calculations to investigate the adsorption of the DOPA derivatives on graphite. The presence of an electron-withdrawing group, such as nitrogen dioxide, strengthens the adsorption by increasing the π-π interaction between DOPA and graphite. To quantify the distribution of electron charge and to gain insights into the charge distribution at interfaces, we performed Bader charge analysis and examined charge density difference plots. We developed a quantitative structure-property relationship (QSPR) model using an artificial neural network (ANN) to predict the adsorption energy. Using the three-dimensional and quantum mechanical electrostatic potential of a molecule as a descriptor, the present quantum NN model shows promising performance as a predictive QSPR model.

Application of Deep Learning for Studying NMDA Receptors

Artificial intelligence underwent remarkable advancement in the past decade, revolutionizing our way of thinking and unlocking unprecedented opportunities across various fields, including drug development. The emergence of large pretrained models, such as ChatGPT, has even begun to demonstrate human-level performance in certain tasks.However, the difficulties of deploying and utilizing AI and pretrained model for nonexpert limited its practical use. To overcome this challenge, here we presented three highly accessible online tools based on a large pretrained model for chemistry, the Uni-Mol, for drug development against CNS diseases, including those targeting NMDA receptor: the blood-brain barrier (BBB) permeability prediction, the quantitative structure-activity relationship (QSAR) analysis system, and a versatile interface of the AI-based molecule generation model named VD-gen. We believe that these resources will effectively bridge the gap between cutting-edge AI technology and NMDAR experts, facilitating rapid and rational drug development.

In silico modeling revealed phytomolecules derived from Cymbopogon citratus (DC.) leaf extract as promising candidates for malaria therapy

The emergence of varying levels of resistance to currently available antimalarial drugs significantly threatens global health. This factor heightens the urgency to explore bioactive compounds from natural products with a view to discovering and developing newer antimalarial drugs with novel mode of actions. Therefore, we evaluated the inhibitory effects of sixteen phytocompounds from Cymbopogon citratus leaf extract against Plasmodium falciparum drug targets such as P. falciparum circumsporozoite protein (PfCSP), P. falciparum merozoite surface protein 1 (PfMSP1) and P. falciparum erythrocyte membrane protein 1 (PfEMP1). In silico approaches including molecular docking, pharmacophore modeling and 3D-QSAR were adopted to analyze the inhibitory activity of the compounds under consideration. The molecular docking results indicated that a compound swertiajaponin from C. citratus exhibited a higher binding affinity (-7.8 kcal/mol) to PfMSP1 as against the standard artesunate-amodiaquine (-6.6 kcal/mol). Swertiajaponin also formed strong hydrogen bond interactions with LYS29, CYS30, TYR34, ASN52, GLY55 and CYS28 amino acid residues. In addition, quercetin another compound from C. citratus exhibited significant binding energies -6.8 and -8.3 kcal/mol with PfCSP and PfEMP1, respectively but slightly lower than the standard artemether-lumefantrine with binding energies of -7.4 kcal/mol against PfCSP and -8.7 kcal/mol against PfEMP1. Overall, the present study provides evidence that swertiajaponin and other phytomolecules from C. citratus have modulatory properties toward P. falciparum drug targets and thus may warrant further exploration in early drug discovery efforts against malaria. Furthermore, these findings lend credence to the folkloric use of C. citratus for malaria treatment.Communicated by Ramaswamy H. Sarma.

4D-QSAR and MIA-QSAR Studies of Aminobenzimidazole Derivatives as Fourth-generation EGFR Inhibitors

BACKGROUND

The epidermal growth factor receptor (EGFR) protein has been intensively studied as a therapeutic target for non-small cell lung cancer (NSCLC). The aminobenzimidazole derivatives as the fourth-generation EGFR inhibitors have achieved promising results and overcame EGFR mutations at C797S, del19 and T790M in NSCLC.

OBJECTIVE

In order to understand the quantitative structure-activity relationship (QSAR) of aminobenzimidazole derivatives as EGFRdel19 T790M C797S inhibitors, the four-dimensional QSAR (4D-QSAR) and multivariate image analysis (MIA-QSAR) have been performed on the data of 45 known aminobenzimidazole derivatives.

METHODS

The 4D-QSAR descriptors were acquired by calculating the association energies between probes and aligned conformational ensemble profiles (CEP), and the regression models were established by partial least squares (PLS). In order to further understand and verify the 4D-QSAR model, MIA-QSAR was constructed by using chemical structure pictures to generate descriptors and PLS regression. Furthermore, the molecular docking and averaged noncovalent interactions (aNCI) analysis were also performed to further understand the interactions between ligands and the EGFR targets, which was in good agreement with the 4D-QSAR model.

RESULTS

The established 4D-QSAR and MIA-QSAR models have strong stability and good external prediction ability.

CONCLUSION

These results will provide theoretical guidance for the research and development of aminobenzimidazole derivatives as new EGFRdel19 T790M C797S inhibitors.

AI-based 3D-QSAR model of FDA-approved repurposed drugs for inhibiting sclerostin

BACKGROUND

Wnt activation promotes bone formation and prevents bone loss. The Wnt pathway antagonist sclerostin and additional anti-sclerostin antibodies were discovered as a result of the development of the monoclonal antibody romosozumab. These monoclonal antibodies greatly increase the risk of cardiac arrest. Three-dimensional quantitative structure-activity relationships (3D-QSAR) predicts biological activities of ligands based on their three-dimensional features by employing powerful chemometric investigations such as artificial neural networks (ANNs) and partial least squares (PLS).

OBJECTIVE

In this study, ligand-receptor interactions were investigated using 3D-QSAR Comparative molecular field analysis (CoMFA). Estimates of steric and electrostatic characteristics in CoMFA are made using Lennard-Jones and Coulomb potentials.

METHODS

To identify the conditions necessary for the activity of these molecules, fifty Food and Drug Administration (FDA)-approved medications were chosen for 3D-QSAR investigations and done by CoMFA. For QSAR analysis, there are numerous tools available. This study employed Open 3D-QSAR for analysis due to its simplicity of use and capacity to produce trustworthy results. Four tools were used for the analysis on this platform: Py-MolEdit, Py-ConfSearch, and Py-CoMFA.

RESULTS

Maps that were generated were used to determine the screen's r2 (Coefficient of Multiple Determinations) value and q2 (correlation coefficient). These numbers must be fewer than 1, suggesting a good, trustworthy model. Cross-validated (q2) 0.532 and conventional (r2) correlation values of 0.969 made the CoMFA model statistically significant. The model showed that hydroxamic acid inhibitors are significantly more sensitive to the steric field than the electrostatic field (70%) (30%). This hypothesis states that steric (43.1%), electrostatic (26.4%), and hydrophobic (20.3%) qualities were important in the design of sclerostin inhibitors.

CONCLUSION

With 3D-QSAR and CoMFA, statistically meaningful models were constructed to predict ligand inhibitory effects. The test set demonstrated the model's robustness. This research may aid in the development of more effective sclerostin inhibitors that are synthesised using FDA-approved medications.

Artificial Intelligence in Drug Discovery: A Bibliometric Analysis and Literature Review

Drug discovery is a complex and iterative process, making it ideal for using artificial intelligence (AI). This paper uses a bibliometric approach to reveal AI's trend and underlying structure in drug discovery (AIDD). A total of 4310 journal articles and reviews indexed in Scopus were analyzed, revealing that AIDD has been rapidly growing over the past two decades, with a significant increase after 2017. The United States, China, and the United Kingdom were the leading countries in research output, with academic institutions, particularly the Chinese Academy of Sciences and the University of Cambridge, being the most productive. In addition, industrial companies, including both pharmaceutical and high-tech ones, also made significant contributions. Additionally, this paper thoroughly discussed the evolution and research frontiers of AIDD, which were uncovered through co-occurrence analyses of keywords using VOSviewer. Our findings highlight that AIDD is an interdisciplinary and promising research field that has the potential to revolutionize drug discovery. The comprehensive overview provided here will be of significant interest to researchers, practitioners, and policy-makers in related fields. The results emphasize the need for continued investment and collaboration in AIDD to accelerate drug discovery, reduce costs, and improve patient outcomes.

Machine learning and traditional QSAR modeling methods: a case study of known PXR activators

Pregnane X receptor (PXR), extensively expressed in human tissues related to digestion and metabolism, is responsible for recognizing and detoxifying diverse xenobiotics encountered by humans. To comprehend the promiscuous nature of PXR and its ability to bind a variety of ligands, computational approaches, viz., quantitative structure-activity relationship (QSAR) models, aid in the rapid dereplication of potential toxicological agents and mitigate the number of animals used to establish a meaningful regulatory decision. Recent advancements in machine learning techniques accommodating larger datasets are expected to aid in developing effective predictive models for complex mixtures (viz., dietary supplements) before undertaking in-depth experiments. Five hundred structurally diverse PXR ligands were used to develop traditional two-dimensional (2D) QSAR, machine-learning-based 2D-QSAR, field-based three-dimensional (3D) QSAR, and machine-learning-based 3D-QSAR models to establish the utility of predictive machine learning methods. Additionally, the applicability domain of the agonists was established to ensure the generation of robust QSAR models. A prediction set of dietary PXR agonists was used to externally-validate generated QSAR models. QSAR data analysis revealed that machine-learning 3D-QSAR techniques were more accurate in predicting the activity of external terpenes with an external validation squared correlation coefficient (R2) of 0.70 versus an R2 of 0.52 in machine-learning 2D-QSAR. Additionally, a visual summary of the binding pocket of PXR was assembled from the field 3D-QSAR models. By developing multiple QSAR models in this study, a robust groundwork for assessing PXR agonism from various chemical backbones has been established in anticipation of the identification of potential causative agents in complex mixtures.

Application of artificial intelligence approaches to predict the metabolism of xenobiotic molecules by human gut microbiome

A highly complex, diverse, and dense community of more than 1,000 different gut bacterial species constitutes the human gut microbiome that harbours vast metabolic capabilities encoded by more than 300,000 bacterial enzymes to metabolise complex polysaccharides, orally administered drugs/xenobiotics, nutraceuticals, or prebiotics. One of the implications of gut microbiome mediated biotransformation is the metabolism of xenobiotics such as medicinal drugs, which lead to alteration in their pharmacological properties, loss of drug efficacy, bioavailability, may generate toxic byproducts and sometimes also help in conversion of a prodrug into its active metabolite. Given the diversity of gut microbiome and the complex interplay of the metabolic enzymes and their diverse substrates, the traditional experimental methods have limited ability to identify the gut bacterial species involved in such biotransformation, and to study the bacterial species-metabolite interactions in gut. In this scenario, computational approaches such as machine learning-based tools presents unprecedented opportunities and ability to predict the gut bacteria and enzymes that can potentially metabolise a candidate drug. Here, we have reviewed the need to identify the gut microbiome-based metabolism of xenobiotics and have provided comprehensive information on the available methods, tools, and databases to address it along with their scope and limitations.

MASSA Algorithm: an automated rational sampling of training and test subsets for QSAR modeling

QSAR models capable of predicting biological, toxicity, and pharmacokinetic properties were widely used to search lead bioactive molecules in chemical databases. The dataset's preparation to build these models has a strong influence on the quality of the generated models, and sampling requires that the original dataset be divided into training (for model training) and test (for statistical evaluation) sets. This sampling can be done randomly or rationally, but the rational division is superior. In this paper, we present MASSA, a Python tool that can be used to automatically sample datasets by exploring the biological, physicochemical, and structural spaces of molecules using PCA, HCA, and K-modes. The proposed algorithm is very useful when the variables used for QSAR are not available or to construct multiple QSAR models with the same training and test sets, producing models with lower variability and better values for validation metrics. These results were obtained even when the descriptors used in the QSAR/QSPR were different from those used in the separation of training and test sets, indicating that this tool can be used to build models for more than one QSAR/QSPR technique. Finally, this tool also generates useful graphical representations that can provide insights into the data.

PeptideBERT: A Language Model Based on Transformers for Peptide Property Prediction

Recent advances in language models have enabled the protein modeling community with a powerful tool that uses transformers to represent protein sequences as text. This breakthrough enables a sequence-to-property prediction for peptides without relying on explicit structural data. Inspired by the recent progress in the field of large language models, we present PeptideBERT, a protein language model specifically tailored for predicting essential peptide properties such as hemolysis, solubility, and nonfouling. The PeptideBERT utilizes the ProtBERT pretrained transformer model with 12 attention heads and 12 hidden layers. Through fine-tuning the pretrained model for the three downstream tasks, our model is state of the art (SOTA) in predicting hemolysis, which is crucial for determining a peptide's potential to induce red blood cells as well as nonfouling properties. Leveraging primarily shorter sequences and a data set with negative samples predominantly associated with insoluble peptides, our model showcases remarkable performance.

The Bigger Fish: A Comparison of Meta-Learning QSAR Models on Low-Resourced Aquatic Toxicity Regression Tasks

Toxicological information as needed for risk assessments of chemical compounds is often sparse. Unfortunately, gathering new toxicological information experimentally often involves animal testing. Simulated alternatives, e.g., quantitative structure-activity relationship (QSAR) models, are preferred to infer the toxicity of new compounds. Aquatic toxicity data collections consist of many related tasks─each predicting the toxicity of new compounds on a given species. Since many of these tasks are inherently low-resource, i.e., involve few associated compounds, this is challenging. Meta-learning is a subfield of artificial intelligence that can lead to more accurate models by enabling the utilization of information across tasks. In our work, we benchmark various state-of-the-art meta-learning techniques for building QSAR models, focusing on knowledge sharing between species. Specifically, we employ and compare transformational machine learning, model-agnostic meta-learning, fine-tuning, and multi-task models. Our experiments show that established knowledge-sharing techniques outperform single-task approaches. We recommend the use of multi-task random forest models for aquatic toxicity modeling, which matched or exceeded the performance of other approaches and robustly produced good results in the low-resource settings we studied. This model functions on a species level, predicting toxicity for multiple species across various phyla, with flexible exposure duration and on a large chemical applicability domain.

Potential for Machine Learning to Address Data Gaps in Human Toxicity and Ecotoxicity Characterization

Machine Learning (ML) is increasingly applied to fill data gaps in assessments to quantify impacts associated with chemical emissions and chemicals in products. However, the systematic application of ML-based approaches to fill chemical data gaps is still limited, and their potential for addressing a wide range of chemicals is unknown. We prioritized chemical-related parameters for chemical toxicity characterization to inform ML model development based on two criteria: (1) each parameter's relevance to robustly characterize chemical toxicity described by the uncertainty in characterization results attributable to each parameter and (2) the potential for ML-based approaches to predict parameter values for a wide range of chemicals described by the availability of chemicals with measured parameter data. We prioritized 13 out of 38 parameters for developing ML-based approaches, while flagging another nine with critical data gaps. For all prioritized parameters, we performed a chemical space analysis to assess further the potential for ML-based approaches to predict data for diverse chemicals considering the structural diversity of available measured data, showing that ML-based approaches can potentially predict 8-46% of marketed chemicals based on 1-10% with available measured data. Our results can systematically inform future ML model development efforts to address data gaps in chemical toxicity characterization.

Advancing Computational Toxicology by Interpretable Machine Learning

Chemical toxicity evaluations for drugs, consumer products, and environmental chemicals have a critical impact on human health. Traditional animal models to evaluate chemical toxicity are expensive, time-consuming, and often fail to detect toxicants in humans. Computational toxicology is a promising alternative approach that utilizes machine learning (ML) and deep learning (DL) techniques to predict the toxicity potentials of chemicals. Although the applications of ML- and DL-based computational models in chemical toxicity predictions are attractive, many toxicity models are "black boxes" in nature and difficult to interpret by toxicologists, which hampers the chemical risk assessments using these models. The recent progress of interpretable ML (IML) in the computer science field meets this urgent need to unveil the underlying toxicity mechanisms and elucidate the domain knowledge of toxicity models. In this review, we focused on the applications of IML in computational toxicology, including toxicity feature data, model interpretation methods, use of knowledge base frameworks in IML development, and recent applications. The challenges and future directions of IML modeling in toxicology are also discussed. We hope this review can encourage efforts in developing interpretable models with new IML algorithms that can assist new chemical assessments by illustrating toxicity mechanisms in humans.

A chemical structure and machine learning approach to assess the potential bioactivity of endogenous metabolites and their association with early-childhood hs-CRP levels

Metabolomics has gained much attraction due to its potential to reveal molecular disease mechanisms and present viable biomarkers. In this work we used a panel of untargeted serum metabolomes in 602 childhood patients of the COPSAC2010 mother-child cohort. The annotated part of the metabolome consists of 493 chemical compounds curated using automated procedures. Using predicted quantitative-structure-bioactivity relationships for the Tox21 database on nuclear receptors and stress response in cell lines, we created a filtering method for the vast number of quantified metabolites. The metabolites measured in children's serums used here have predicted potential against the chosen target modelled targets. The targets from Tox21 have been used with quantitative structure-activity relationships (QSARs) and were trained for ~7000 structures, saved as models, and then applied to 493 metabolites to predict their potential bioactivities. The models were selected based on strict accuracy criteria surpassing random effects. After application, 52 metabolites showed potential bioactivity based on structural similarity with known active compounds from the Tox21 set. The filtered compounds were subsequently used and weighted by their bioactive potential to show an association with early childhood hs-CRP levels at six months in a linear model supporting a physiological adverse effect on systemic low-grade inflammation. The significant metabolites were reported.

Data-driven discovery of innate immunomodulators via machine learning-guided high throughput screening

The innate immune response is vital for the success of prophylactic vaccines and immunotherapies. Control of signaling in innate immune pathways can improve prophylactic vaccines by inhibiting unfavorable systemic inflammation and immunotherapies by enhancing immune stimulation. In this work, we developed a machine learning-enabled active learning pipeline to guide in vitro experimental screening and discovery of small molecule immunomodulators that improve immune responses by altering the signaling activity of innate immune responses stimulated by traditional pattern recognition receptor agonists. Molecules were tested by in vitro high throughput screening (HTS) where we measured modulation of the nuclear factor κ-light-chain-enhancer of activated B-cells (NF-κB) and the interferon regulatory factors (IRF) pathways. These data were used to train data-driven predictive models linking molecular structure to modulation of the NF-κB and IRF responses using deep representational learning, Gaussian process regression, and Bayesian optimization. By interleaving successive rounds of model training and in vitro HTS, we performed an active learning-guided traversal of a 139 998 molecule library. After sampling only ∼2% of the library, we discovered viable molecules with unprecedented immunomodulatory capacity, including those capable of suppressing NF-κB activity by up to 15-fold, elevating NF-κB activity by up to 5-fold, and elevating IRF activity by up to 6-fold. We extracted chemical design rules identifying particular chemical fragments as principal drivers of specific immunomodulation behaviors. We validated the immunomodulatory effect of a subset of our top candidates by measuring cytokine release profiles. Of these, one molecule induced a 3-fold enhancement in IFN-β production when delivered with a cyclic di-nucleotide stimulator of interferon genes (STING) agonist. In sum, our machine learning-enabled screening approach presents an efficient immunomodulator discovery pipeline that has furnished a library of novel small molecules with a strong capacity to enhance or suppress innate immune signaling pathways to shape and improve prophylactic vaccination and immunotherapies.

Harnessing Semi-Supervised Machine Learning to Automatically Predict Bioactivities of Per- and Polyfluoroalkyl Substances (PFASs)

Many per- and polyfluoroalkyl substances (PFASs) pose significant health hazards due to their bioactive and persistent bioaccumulative properties. However, assessing the bioactivities of PFASs is both time-consuming and costly due to the sheer number and expense of in vivo and in vitro biological experiments. To this end, we harnessed new unsupervised/semi-supervised machine learning models to automatically predict bioactivities of PFASs in various human biological targets, including enzymes, genes, proteins, and cell lines. Our semi-supervised metric learning models were used to predict the bioactivity of PFASs found in the recent Organisation of Economic Co-operation and Development (OECD) report list, which contains 4730 PFASs used in a broad range of industries and consumers. Our work provides the first semi-supervised machine learning study of structure-activity relationships for predicting possible bioactivities in a variety of PFAS species.

Evaluating uncertainty-based active learning for accelerating the generalization of molecular property prediction

Deep learning models have proven to be a powerful tool for the prediction of molecular properties for applications including drug design and the development of energy storage materials. However, in order to learn accurate and robust structure-property mappings, these models require large amounts of data which can be a challenge to collect given the time and resource-intensive nature of experimental material characterization efforts. Additionally, such models fail to generalize to new types of molecular structures that were not included in the model training data. The acceleration of material development through uncertainty-guided experimental design has the promise to significantly reduce the data requirements and enable faster generalization to new types of materials. To evaluate the potential of such approaches for electrolyte design applications, we perform comprehensive evaluation of existing uncertainty quantification methods on the prediction of two relevant molecular properties - aqueous solubility and redox potential. We develop novel evaluation methods to probe the utility of the uncertainty estimates for both in-domain and out-of-domain data sets. Finally, we leverage selected uncertainty estimation methods for active learning to evaluate their capacity to support experimental design.

Usage of model combination in computational toxicology

New Approach Methodologies (NAMs) have ushered in a new era in the field of toxicology, aiming to replace animal testing. However, despite these advancements, they are not exempt from the inherent complexities associated with the study's endpoint. In this review, we have identified three major groups of complexities: mechanistic, chemical space, and methodological. The mechanistic complexity arises from interconnected biological processes within a network that are challenging to model in a single step. In the second group, chemical space complexity exhibits significant dissimilarity between compounds in the training and test series. The third group encompasses algorithmic and molecular descriptor limitations and typical class imbalance problems. To address these complexities, this work provides a guide to the usage of a combination of predictive Quantitative Structure-Activity Relationship (QSAR) models, known as metamodels. This combination of low-level models (LLMs) enables a more precise approach to the problem by focusing on different sub-mechanisms or sub-processes. For mechanistic complexity, multiple Molecular Initiating Events (MIEs) or levels of information are combined to form a mechanistic-based metamodel. Regarding the complexity arising from chemical space, two types of approaches were reviewed to construct a fragment-based chemical space metamodel: those with and without structure sharing. Metamodels with structure sharing utilize unsupervised strategies to identify data patterns and build low-level models for each cluster, which are then combined. For situations without structure sharing due to pharmaceutical industry intellectual property, the use of prediction sharing, and federated learning approaches have been reviewed. Lastly, to tackle methodological complexity, various algorithms are combined to overcome their limitations, diverse descriptors are employed to enhance problem definition and balanced dataset combinations are used to address class imbalance issues (methodological-based metamodels). Remarkably, metamodels consistently outperformed classical QSAR models across all cases, highlighting the importance of alternatives to classical QSAR models when faced with such complexities.

Artificial intelligence-guided approach for efficient virtual screening of hits against Schistosoma mansoni

Background: The impact of schistosomiasis, which affects over 230 million people, emphasizes the urgency of developing new antischistosomal drugs. Artificial intelligence is vital in accelerating the drug discovery process. Methodology & results: We developed classification and regression machine learning models to predict the schistosomicidal activity of compounds not experimentally tested. The prioritized compounds were tested on schistosomula and adult stages of Schistosoma mansoni. Four compounds demonstrated significant activity against schistosomula, with 50% effective concentration values ranging from 9.8 to 32.5 μM, while exhibiting no toxicity in animal and human cell lines. Conclusion: These findings represent a significant step forward in the discovery of antischistosomal drugs. Further optimization of these active compounds can pave the way for their progression into preclinical studies.

Stacked ensemble learning on HaCaT cytotoxicity for skin irritation prediction: A case study on dipterocarpol

Skin irritation is an adverse effect associated with various substances, including chemicals, drugs, or natural products. Dipterocarpol, extracted from Dipterocarpus alatus, contains several skin benefits notably anticancer, wound healing, and antibacterial properties. However, the skin irritation of dipterocarpol remains unassessed. Quantitative structure-activity relationship (QSAR) is a recommended tool for toxicity assessment involving less time, money, and animal testing to access unavailable acute toxicity data. Therefore, our study aimed to develop a highly accurate machine learning-based QSAR model for predicting skin irritation. We utilized a stacked ensemble learning model with 1064 chemicals. We also adhered to the recommendations from the OECD for QSAR validation. Subsequently, we used the proposed model to explore the cytotoxicity of dipterocarpol on keratinocytes. Our findings indicate that the model displayed promising statistical quality in terms of accuracy, precision, and recall in both 10-fold cross-validation and test datasets. Moreover, the model predicted that dipterocarpol does not have skin irritation, which was confirmed by the cell-based assay. In conclusion, our proposed model can be applied for the risk assessment of skin irritation in untested compounds that fall within its applicability domain. The web application of this model is available at https://qsarlabs.com/#stackhacat.

In silico QSAR modeling to predict the safe use of antibiotics during pregnancy

The use of medicines during pregnancy is a growing public health concern due to the risk of developmental toxicity. Healthcare providers heavily rely on the FDA pregnancy risk categories (A, B, C, D, and X). Antibiotics are among the most prescribed drugs during pregnancy and are often listed under category B or C. However, the risk-benefit assessment may be lacking due to challenges in the clinical toxicology studies on pregnant women, such as ethical concerns. The primary focus of this study is to generate a model that predicts the safe use of antibiotics during pregnancy by using in silico approaches. Thus, a QSAR model was created to assess the FDA pregnancy category (B or C) of antibiotics. The dataset consisted of 97 antibiotics obtained from the FDA. A total of 6420 molecular descriptors were determined via multiple software and various machine learning algorithms were utilized. The performance of the models was measured using internal and external validation. The accuracy (ACC) values of the most successful model were 83.82% for the internal and 94.11% for the external validation. Sensitivity (SE), specificity (SP), MCC, and ROC values were 0.878, 0.778, 0.68, and 0.892 for the internal validation and 0.9, 1, 0.887, and 0.936 for the external validation, respectively. Kappa statistics also indicate that there was a substantial agreement for internal validation with 0.6765 and an almost perfect agreement for external validation with 0.8811. In conclusion, our model can be used as an initial step before pre-clinical and clinical studies to predict the safe use of antibiotics in pregnancy.