A review on machine learning approaches and trends in drug discovery

Harnessing machine learning in contemporary tobacco research

Machine learning (ML) has the potential to transform tobacco research and address the urgent public health crisis posed by tobacco use. Despite the well-documented health risks, cessation rates remain low. ML techniques offer innovative solutions by analyzing vast datasets to uncover patterns in smoking behavior, genetic predispositions, and effective cessation strategies. ML can predict smoking-induced non-communicable diseases (SiNCDs) like lung cancer and postmenopausal osteoporosis by identifying biomarkers and genetic profiles, generating personalized predictions, and guiding interventions. It also improves prediction of infant tobacco smoke exposure, distinguishes secondhand and thirdhand smoke, and enhances protection strategies for children. Data-driven, personalized approaches using ML track real-time data for personalized feedback and offer timely interventions, continuously improving cessation strategies. Overall, ML provides sophisticated predictive models, enhances understanding of complex biological mechanisms, and enables personalized interventions, demonstrating significant potential in the fight against the tobacco epidemic.

Machine Learning for Toxicity Prediction Using Chemical Structures: Pillars for Success in the Real World

Machine learning (ML) is increasingly valuable for predicting molecular properties and toxicity in drug discovery. However, toxicity-related end points have always been challenging to evaluate experimentally with respect to in vivo translation due to the required resources for human and animal studies; this has impacted data availability in the field. ML can augment or even potentially replace traditional experimental processes depending on the project phase and specific goals of the prediction. For instance, models can be used to select promising compounds for on-target effects or to deselect those with undesirable characteristics (e.g., off-target or ineffective due to unfavorable pharmacokinetics). However, reliance on ML is not without risks, due to biases stemming from nonrepresentative training data, incompatible choice of algorithm to represent the underlying data, or poor model building and validation approaches. This might lead to inaccurate predictions, misinterpretation of the confidence in ML predictions, and ultimately suboptimal decision-making. Hence, understanding the predictive validity of ML models is of utmost importance to enable faster drug development timelines while improving the quality of decisions. This perspective emphasizes the need to enhance the understanding and application of machine learning models in drug discovery, focusing on well-defined data sets for toxicity prediction based on small molecule structures. We focus on five crucial pillars for success with ML-driven molecular property and toxicity prediction: (1) data set selection, (2) structural representations, (3) model algorithm, (4) model validation, and (5) translation of predictions to decision-making. Understanding these key pillars will foster collaboration and coordination between ML researchers and toxicologists, which will help to advance drug discovery and development.

Identification of potential 3CLpro inhibitors-modulators for human norovirus infections through an advanced virtual screening approach

The present study aimed to screen small molecular compounds such as human noroviruses (HuNoV) inhibitors/modulators that could potentially be responsible for exhibiting some magnitude of inhibitory/modulatory activity against HuNoV 3CLPro. The structural similarity-based screening against the ChEMBL database is performed against known chemical entities that are presently under pre-clinical trial. After the similarity search, remaining molecules were considered for molecular docking using SCORCH and PLANTS. On detailed analyses and comparisons with the control molecule, three hits (CHEMBL393820, CHEMBL2028556, and CHEMBL3747799) were found to have the potential for HuNoV 3CLpro inhibition/modulation. The binding interaction analysis revealed several critical amino acids responsible to hold the molecules tightly at the close proximity site of the catalytic residues of HuNoV 3CLpro. Further, MD simulation study was performed in triplicate to understand the binding stability and potentiality of the proposed molecule toward HuNov 3CLpro. The binding free energy based on MM-GBSA has revealed their strong interaction affinity with 3CLpro.

The Role of Artificial Intelligence in Drug Discovery and Pharmaceutical Development: A Paradigm Shift in the History of Pharmaceutical Industries

In today's world, with an increasing patient population, the need for medications is increasing rapidly. However, the current practice of drug development is time-consuming and requires a lot of investment by the pharmaceutical industries. Currently, it takes around 8-10 years and $3 billion of investment to develop a medication. Pharmaceutical industries and regulatory authorities are continuing to adopt new technologies to improve the efficiency of the drug development process. However, over the decades the pharmaceutical industries were not able to accelerate the drug development process. The pandemic (COVID-19) has taught the pharmaceutical industries and regulatory agencies an expensive lesson showing the need for emergency preparedness by accelerating the drug development process. Over the last few years, the pharmaceutical industries have been collaborating with artificial intelligence (AI) companies to develop algorithms and models that can be implemented at various stages of the drug development process to improve efficiency and reduce the developmental timelines significantly. In recent years, AI-screened drug candidates have entered clinical testing in human subjects which shows the interest of pharmaceutical companies and regulatory agencies. End-end integration of AI within the drug development process will benefit the industries for predicting the pharmacokinetic and pharmacodynamic profiles, toxicity, acceleration of clinical trials, study design, virtual monitoring of subjects, optimization of manufacturing process, analyzing and real-time monitoring of product quality, and regulatory preparedness. This review article discusses in detail the role of AI in various avenues of the pharmaceutical drug development process, its limitations, regulatory and future perspectives.

Machine Learning for Prediction of Drug Concentrations: Application and Challenges

With the advancements in algorithms and increased accessibility of multi-source data, machine learning in pharmacokinetics is gaining interest. This review summarizes studies on machine learning-based pharmacokinetics analysis up to September 2024, identified from the PubMed and IEEE Xplore databases. The main focus of this review is on the use of machine learning in predicting drug concentration. This review provides a comprehensive summary of the advances in the machine learning algorithms for pharmacokinetics analysis. Specifically, we describe the common practices in data preprocessing, the application scenarios of various algorithms, and the critical challenges that require attention. Most machine learning models show comparable performance to those of population pharmacokinetics models. Tree-based algorithms and neural networks have the most applications. Furthermore, the use of ensemble modeling techniques can improve the accuracy of these models' predictions of drug concentrations, especially the ensembles of machine learning and pharmacometrics.

Significance of NMDA receptor-targeting compounds in neuropsychological disorders: An in-depth review

N-methyl-D-aspartate receptors (NMDARs), a subclass of glutamate-gated ion channels, play an integral role in the maintenance of synaptic plasticity and excitation-inhibition balance within the central nervous system (CNS). Any irregularities in NMDAR functions, whether hypo-activation or over-activation, can destabilize neural networks and impair CNS function. Several decades of experimental and clinical investigations have demonstrated that NMDAR dysfunction is implicated in the pathophysiology of various neurological disorders. Despite designing a long list of compounds that differentially modulate NMDARs, success in developing drugs that can selectively and effectively regulate various NMDAR subtypes while showing encouraging efficacy in clinical settings remains limited. A better understanding of the basic mechanism of NMDAR function, particularly its selective regulation in pathological conditions, could aid in designing effective drugs for the treatment of neurological conditions. Here, we reviewed the experimental and clinical investigations that studied the effects of available NMDAR modulators in various neurological disorders and weighed up the pros and cons of the use of these substances on the improvement of functional outcomes of these disorders. Despite numerous efforts to develop NMDAR modulatory drugs that did not produce the desired outcomes, NMDARs remain a significant target for advancing novel drugs to treat neurological disorders. This article reviews the complexity of NMDAR signaling dysfunction in different neurological diseases, the efforts taken to examine designed compounds targeting specific subtypes of NMDARs, including challenges associated with using these substances, and the potential enhancements in drug discovery for NMDAR modulatory compounds by innovative technologies.

Machine learning: Python tools for studying biomolecules and drug design

The increasing adoption of computational methods and artificial intelligence in scientific research has led to a growing interest in versatile tools like Python. In the fields of medical chemistry, biochemistry, and bioinformatics, Python has emerged as a key language for tackling complex challenges. It is used to solve various tasks, such as drug discovery, high-throughput and virtual screening, protein and genome analysis, and predicting drug efficacy. This review presents a list of tools for these tasks, including scripts, libraries, and ready-made programs, and serves as a starting point for scientists wishing to apply automation or optimization to routine tasks in medical chemistry and bioinformatics.

The physics-AI dialogue in drug design

A long path has led from the determination of the first protein structure in 1960 to the recent breakthroughs in protein science. Protein structure prediction and design methodologies based on machine learning (ML) have been recognized with the 2024 Nobel prize in Chemistry, but they would not have been possible without previous work and the input of many domain scientists. Challenges remain in the application of ML tools for the prediction of structural ensembles and their usage within the software pipelines for structure determination by crystallography or cryogenic electron microscopy. In the drug discovery workflow, ML techniques are being used in diverse areas such as scoring of docked poses, or the generation of molecular descriptors. As the ML techniques become more widespread, novel applications emerge which can profit from the large amounts of data available. Nevertheless, it is essential to balance the potential advantages against the environmental costs of ML deployment to decide if and when it is best to apply it. For hit to lead optimization ML tools can efficiently interpolate between compounds in large chemical series but free energy calculations by molecular dynamics simulations seem to be superior for designing novel derivatives. Importantly, the potential complementarity and/or synergism of physics-based methods (e.g., force field-based simulation models) and data-hungry ML techniques is growing strongly. Current ML methods have evolved from decades of research. It is now necessary for biologists, physicists, and computer scientists to fully understand advantages and limitations of ML techniques to ensure that the complementarity of physics-based methods and ML tools can be fully exploited for drug design.

Classification and regression machine learning models for predicting mixed toxicity of carbamazepine and its transformation products

Carbamazepine (CBZ) and its transformation products (TPs) often occur in aquatic environments in the form of mixtures, posing potential risks to ecosystems. However, establishing standardized protocols for synthesizing, isolating, and acquiring these TPs has been challenging, leading to difficulty in obtaining toxicity data. Accurately assessing the risks associated with mixed toxicity of TPs was therefore critical. The research evaluated the binary toxicity of CBZ and its TPs using luminescent bacteria. The mixed toxicity of TPs showed simply additive effects. In order to comprehend the connection between the toxicity of TPs and CBZ, we labeled TPs with toxicity higher than CBZ as 'high risk' and TPs with lower toxicity as 'low risk.' Subsequently, we developed and tested seven classification models and five regression models. The classification models were of guiding significance for the management of toxicity risk. In contrast, the regression models are capable of addressing the lack of mixed toxicity data. However, these regression models show limitations on the experimental datasets, and their performance is unsatisfactory. Considering the challenges in obtaining toxicity data of transformed products, we addressed this limitation by enhancing the dataset using generative adversarial networks (GANs), thereby improving the generalization capability about the regression models. This study highlighted the potential of quantitative structure-activity relationship (QSAR) models, which are based on based on machine learning for predicting the mixed toxicity of TPs, providing a solution for toxicity assessment without chemical standards.

Meta-analysis and review of in silico methods in drug discovery - part 1: technological evolution and trends from big data to chemical space

This review offers an overview of advanced in silico methods crucial for drug discovery, emphasizing their integration with data science, and investigates the effectiveness of data science, machine learning, and artificial intelligence via a thorough meta-analysis of existing technologies. This meta-analysis aims to rank these technologies based on their applications and accessibility of knowledge. Initially, a search strategy yielded 900 papers, which were then refined into two subsets: the top 300 most-cited papers since 2000 and papers selected for systematic review based on high impact. From these, 97 articles were identified for discussion, categorized by their influence on society. The focus remains on the qualitative impact of these disciplines rather than solely on metrics like new drug approvals. Ultimately, the review underscores the role of big data in enhancing our comprehension of drug candidate trajectories from development to commercialization, utilizing information stored in publicly available databases to chemical space. Graphical extrapolation of some keywords (Drug Discovery; Big Data; Database; Metadata) discussed in this article and their evolution (in terms of absolute items that are available) by time.

Deciphering Ferroptosis: From Molecular Pathways to Machine Learning-Guided Therapeutic Innovation

Ferroptosis is a unique form of cell death reliant on iron and lipid peroxidation. It disrupts redox balance, causing cell death by damaging the plasma membrane, with inducers acting through enzymatic pathways or transport systems. In cancer treatment, suppressing ferroptosis or circumventing it holds significant promise. Beyond cancer, ferroptosis affects aging, organs, metabolism, and nervous system. Understanding ferroptosis mechanisms holds promise for uncovering novel therapeutic strategies across a spectrum of diseases. However, detection and regulation of this regulated cell death are still mired with challenges. The dearth of cell, tissue, or organ-specific biomarkers muted the pharmacological use of ferroptosis. This review covers recent studies on ferroptosis, detailing its properties, key genes, metabolic pathways, and regulatory networks, emphasizing the interaction between cellular signaling and ferroptotic cell death. It also summarizes recent findings on ferroptosis inducers, inhibitors, and regulators, highlighting their potential therapeutic applications across diseases. The review addresses challenges in utilizing ferroptosis therapeutically and explores the use of machine learning to uncover complex patterns in ferroptosis-related data, aiding in the discovery of biomarkers, predictive models, and therapeutic targets. Finally, it discusses emerging research areas and the importance of continued investigation to harness the full therapeutic potential of targeting ferroptosis.

Enhancing HCV NS3 Inhibitor Classification with Optimized Molecular Fingerprints Using Random Forest

The classification of Hepatitis C virus (HCV) NS3 inhibitors is essential for identifying potential antiviral agents through computational methods. This study aims to develop an optimized machine learning (ML) model using random forest (RF) and molecular fingerprints to accurately classify HCV NS3 inhibitors. A dataset of 965 molecules was retrieved from the ChEMBL database, and 290 bioactive compounds were selected for model training. Twelve molecular fingerprint descriptors were tested, and the CDK graph-only fingerprint yielded the best performance. In addition to RF, performance comparisons of other classifiers such as instance-based k-nearest neighbor (IBk), logistic regression (LR), AdaBoost, and OneR were conducted using WEKA with various molecular fingerprint descriptors. The optimized RF model achieved an accuracy of 89.6552%, a mean absolute error (MAE) of 0.2114, a root mean square error (RMSE) of 0.3304, and a Matthews correlation coefficient (MCC) of 0.7950 on the test set. These results highlight the effectiveness of optimized molecular fingerprints in enhancing virtual screening (VS) for HCV inhibitors. This approach offers a data-driven method for drug discovery.

Machine learning-based prediction models for renal impairment in Chinese adults with hyperuricaemia: risk factor analysis

In hyperuricaemic populations, multiple factors may contribute to impaired renal function. This study aimed to establish a machine learning-based model to identify characteristic factors related to renal impairment in hyperuricaemic patients, determine dose‒response relationships, and facilitate early intervention strategies. Data were collected through the big data platform of Nanjing Hospital of Traditional Chinese Medicine, encompassing 2,705 patients with hyperuricaemia (1,577 with renal impairment, 828 without) from June 2019 to June 2022. After multiple imputations for missing values, the dataset was randomly split into training (70%) and validation (30%) sets. We employed three machine learning algorithms for feature selection: random forest (with 100 decision trees and an OOB error rate of 23.34%), LASSO regression (optimal lambda of -3.58), and XGBoost (learning rate of 0.3, maximum tree depth of 1, and 50 rounds of boosting). The intersection of features identified by these algorithms through Venn diagram analysis yielded four key predictors. A logistic regression model was subsequently constructed and evaluated for discrimination (AUC), calibration (Brier score), and clinical utility (DCA). Restricted cubic spline (RCS) curves were utilized to analyse the dose‒response relationships. The model, which incorporates age, cystatin C (Cys-C), uric acid (UA), and sex, demonstrated robust performance, with an AUC of 0.818 [95% CI (0.796-0.817)] in the training set and an AUC of 0.82 [95% CI (0.787-0.853)] in the validation set. Calibration tests yielded Brier scores of 0.160 and 0.158, respectively. Clinical decision curves revealed optimal prediction probability intervals of 6-99.02% and 7-93.14%. In the hyperuricaemic population, each 0.5 mg/L increase in Cys-C, 10-year increase in age, and 100 µmol/L increase in UA corresponded to increased risks of 13%, 81%, and 73%, respectively. RCS analysis revealed nonlinear relationships for Age and Cys-C and a linear relationship for UA, with sex-specific distribution patterns. The machine learning-based model incorporating these four indicators demonstrated excellent predictive performance for renal impairment in hyperuricaemic patients. These findings suggest that monitoring Cys-C and UA levels while considering age and sex differences is crucial for risk assessment and prevention strategies.

Innovative perspectives on the discovery of small molecule antibiotics

Antibiotics are essential to modern medicine, but multidrug-resistant (MDR) bacterial infections threaten their efficacy. Resistance evolution shortens antibiotic lifespans, limiting investment returns and slowing new approvals. Consequently, the WHO defines four innovation criteria: new chemical class, target, mode of action (MoA), and lack of cross-resistance. This review explores innovative discovery approaches, including AI-driven screening, metagenomics, and target-based strategies, to develop novel antibiotics that meet these criteria and combat MDR infections.

Predicting Antimicrobial Class Specificity of Small Molecules Using Machine Learning

While the useful armory of antibiotic drugs is continually depleted due to the emergence of drug-resistant pathogens, the development of novel therapeutics has also slowed down. In the era of advanced computational methods, approaches like machine learning (ML) could be one potential solution to help reduce the high costs and complexity of antibiotic drug discovery and attract collaboration across organizations. In our work, we developed a large antimicrobial knowledge graph (AntiMicrobial-KG) as a repository for collecting and visualizing public in vitro antibacterial assay. Utilizing this data, we build ML models to efficiently scan compound libraries to identify compounds with the potential to exhibit antimicrobial activity. Our strategy involved training seven classic ML models across six compound fingerprint representations, of which the Random Forest trained on the MHFP6 fingerprint outperformed, demonstrating an accuracy of 75.9% and Cohen's Kappa score of 0.68. Finally, we illustrated the model's applicability for predicting the antimicrobial properties of two small molecule screening libraries. First, the EU-OpenScreen library was tested against a panel of Gram-positive, Gram-negative, and Fungal pathogens. Here, we unveiled that the model was able to correctly predict more than 30% of active compounds for Gram-positive, Gram-negative, and Fungal pathogens. Second, with the Enamine library, a commercially available HTS compound collection with claimed antibacterial properties, we predicted its antimicrobial activity and pathogen class specificity. These results may provide a means for accelerating research in AMR drug discovery efforts by carefully filtering out compounds from commercial libraries with lower chances of being active.

Unveiling the effect of urinary xenoestrogens on chronic kidney disease in adults: A machine learning model

Exposure to three primary xenoestrogens (XEs), including phthalates, parabens, and phenols, has been strongly associated with chronic kidney disease (CKD). An interpretable machine learning (ML) model was developed to predict CKD using data from the National Health and Nutrition Examination Survey (NHANES) database spanning from 2007 to 2016. Four ML algorithms-random forest classifier (RF), XGBoost (XGB), k-nearest neighbors (KNN), and support vector machine (SVM)-were used alongside traditional logistic regression to predict CKD. The study included 6910 U.S. adults, with XGB showing the highest predictive accuracy, achieving an area under the curve (AUC) of 0.817 (95 % CI: 0.789, 0.844). The selected model was interpreted using Shapley additive explanations (SHAP) and partial dependence plot (PDP). The SHAP method identified key predictive features for CKD in urinary metabolites of XEs-methyl paraben (MeP), mono-(carboxynonyl) phthalate (MCNP), and triclosan (TCS)-and suggested personalized CKD care should focus on XE control. PDP results confirmed that, within certain ranges, MeP levels positively impacted the model, MCNP levels negatively impacted it, and TCS had a mixed effect. The synergistic effects suggested that managing urinary MeP levels could be essential for the effective control of CKD. In summary, our research highlights the significant predictive potential of XEs for CKD, especially MeP, MCNP, and TCS.

Prediction of cancer cell line-specific synergistic drug combinations based on multi-omics data

Compared to single-drug therapy, combination therapy involves the use of two or more drugs to reduce drug dosage, decrease drug toxicity, and improve treatment efficacy. We developed an extreme gradient boosting (XGBoost)-based drug-drug cell line prediction model (XDDC) to predict synergistic drug combinations. XDDC was based on XGBoost and used one of the largest drug combination datasets, NCI-ALMANAC. In XDDC, drug chemical structures, adverse drug reactions, and target information were selected as drug features; gene expression, methylation, mutations, copy number variations, and RNA interference data were used as cell line features; and pathway information was incorporated to link drug features and cell line features. XDDC improved the interpretability of drug combination features and outperformed other machine learning methods. It achieved an area under the curve (AUC) of 0.966 ± 0.002 and an AUPR of 0.957 ± 0.002 when cross-validated on NCI-ALMANAC data. Different types of omics data were evaluated and compared in the model. Literature and experimental verification confirmed some of our predictions. XDDC could help medical professionals to rapidly screen synergistic drug combinations against specific cancer cell lines.

Explainable AI-driven prediction of APE1 inhibitors: enhancing cancer therapy with machine learning models and feature importance analysis

The viability of cells and the integrity of the genome depend on the detection and repair of damaged DNA through intricate mechanisms. Cancer treatment employs chemotherapy or radiation therapy to eliminate neoplastic cells by causing substantial damage to their DNA. In many cases, improved DNA repair mechanisms lead to resistance to these medicines; therefore, it is essential to expand efforts to develop drugs that can sensitise cells to these treatments by inhibiting the DNA repair process. Multiple studies have demonstrated a correlation between the overexpression of Apurinic/Apyrimidinic Endonuclease (APE1), the primary mammalian enzyme responsible for excising apurinic or apyrimidinic sites in DNA, and the resistance of cells to cancer therapies; in contrast, APE1 downregulation increases cellular susceptibility to DNA-damaging agents. Thus, the effectiveness of existing therapies can be improved by promoting the targeted sensitization of cancer cells while protecting healthy cells. The current study aims to employ explainable artificial intelligence (XAI) to enhance the accuracy and reliability of machine learning models for the prediction of APE1 inhibitors. Various ML-based regression models are employed to predict the pIC50 value of different medicines. Bayesian optimization and the Permutation Feature Importance (PFI) approach are employed to determine the best hyperparameters of machine learning models and to discover the most significant features for recognizing drug candidates that target APE1 enzymes, respectively. To acquire comprehensive elucidations for the predictive models in our research, two XAI methodologies, namely SHAP and LIME, are used. The SHAP analysis reveals that the features 'C1SP2' and 'ASP-2' are essential in influencing the model's predictions. The SHAP values demonstrate variability for features such as 'maxHBint2' and 'GATS1s,' signifying that their impact is dependent on specific instances within the dataset. The LIME study corroborates these findings, demonstrating that 'C1SP2' and 'ASP-2' are the most significant positive contributors, whereas features like 'SHCHnX,' 'nHdCH2,' and 'GATS1s' result in a decrease in the predicted values. Due to the limited sample size of the APE1 dataset, direct training on this dataset posed challenges in model generalization and reliability. To overcome this limitation, the BACE-1 dataset is leveraged for model training, enabling the ML models to learn from a more extensive and diverse chemical space. Among the tested algorithms, XGBoost demonstrated superior predictive performance, achieving R2 = 0.890, MAE = 0.186, and RMSE = 0.245, significantly surpassing state-of-the-art methods, such as LightGBM and QSAR-ML, which attained R2 scores of 0.798 and 0.630, respectively. These results highlight the robustness of our approach, demonstrating its enhanced generalization capability and superior predictive accuracy compared to existing methodologies.

Computational Modeling of Pharmaceuticals with an Emphasis on Crossing the Blood-Brain Barrier

The discovery and development of new pharmaceutical drugs is a costly, time-consuming, and highly manual process, with significant challenges in ensuring drug bioavailability at target sites. Computational techniques are highly employed in drug design, particularly to predict the pharmacokinetic properties of molecules. One major kinetic challenge in central nervous system drug development is the permeation through the blood-brain barrier (BBB). Several different computational techniques are used to evaluate both BBB permeability and target delivery. Methods such as quantitative structure-activity relationships, machine learning models, molecular dynamics simulations, end-point free energy calculations, or transporter models have pros and cons for drug development, all contributing to a better understanding of a specific characteristic. Additionally, the design (assisted or not by computers) of prodrug and nanoparticle-based drug delivery systems can enhance BBB permeability by leveraging enzymatic activation and transporter-mediated uptake. Neuroactive peptide computational development is also a relevant field in drug design, since biopharmaceuticals are on the edge of drug discovery. By integrating these computational and formulation-based strategies, researchers can enhance the rational design of BBB-permeable drugs while minimizing off-target effects. This review is valuable for understanding BBB selectivity principles and the latest in silico and nanotechnological approaches for improving CNS drug delivery.

dyphAI dynamic pharmacophore modeling with AI: a tool for efficient screening of new acetylcholinesterase inhibitors

Therapeutic strategies for Alzheimer's disease (AD) often involve inhibiting acetylcholinesterase (AChE), underscoring the need for novel inhibitors with high selectivity and minimal side effects. A detailed analysis of the protein-ligand pharmacophore dynamics can facilitate this. In this study, we developed and employed dyphAI, an innovative approach integrating machine learning models, ligand-based pharmacophore models, and complex-based pharmacophore models into a pharmacophore model ensemble. This ensemble captures key protein-ligand interactions, including π-cation interactions with Trp-86 and several π-π interactions with residues Tyr-341, Tyr-337, Tyr-124, and Tyr-72. The protocol identified 18 novel molecules from the ZINC database with binding energy values ranging from -62 to -115 kJ/mol, suggesting their strong potential as AChE inhibitors. To further validate the predictions, nine molecules were acquired and tested for their inhibitory activity against human AChE. Experimental results revealed that molecules, 4 (P-1894047), with its complex multi-ring structure and numerous hydrogen bond acceptors, and 7 (P-2652815), characterized by a flexible, polar framework with ten hydrogen bond donors and acceptors, exhibited IC₅₀ values lower than or equal to that of the control (galantamine), indicating potent inhibitory activity. Similarly, molecules 5 (P-1205609), 6 (P-1206762), 8 (P-2026435), and 9 (P-533735) also demonstrated strong inhibition. In contrast, molecule 3 (P-617769798) showed a higher IC50 value, and molecules 1 (P-14421887) and 2 (P-25746649) yielded inconsistent results, likely due to solubility issues in the experimental setup. These findings underscore the value of integrating computational predictions with experimental validation, enhancing the reliability of virtual screening in the discovery of potent enzyme inhibitors.

Comparative Analysis of pK a Predictions for Arsonic Acids Using Density Functional Theory-Based and Machine Learning Approaches

Arsonic acids (RAsO(OH)2), prevalent in contaminated food, water, air, and soil, pose significant environmental and health risks due to their variable ionization states, which influence key properties such as lipophilicity, solubility, and membrane permeability. Accurate pK a prediction for these compounds is critical yet challenging, as existing models often exhibit limitations across diverse chemical spaces. This study presents a comparative analysis of pK a predictions for arsonic acids using a support vector machine-based machine learning (ML) approach and three density functional theory (DFT)-based models. The DFT models evaluated include correlations to the maximum surface electrostatic potential (V S,max), atomic charges derived from a solvation model (solvation model based on density), and a scaled solvent-accessible surface method. Results indicate that the scaled solvent-accessible surface approach yielded high mean unsigned errors, rendering it less effective. In contrast, the atomic charge-based method on the conjugated arsonate base provided the most accurate predictions. The ML-based approach demonstrated strong predictive performance, suggesting its potential utility in broader chemical spaces. The obtained values for pK a from V S,max show a weak prediction level, because the way of predicting pK a is related only to the electrostatic character of the molecule. However, pK a is influenced by many factors, including the molecular structure, solvation, resonance, inductive effects, and local atomic environments. V S,max cannot fully capture these different interactions, as it gives a simplistic view of the overall molecular potential field.

Therapeutic exploration potential of adenosine receptor antagonists through pharmacophore ligand-based modelling and pharmacokinetics studies against Parkinson disease

Parkinson's Disease (PD) is a neurodegenerative disorder that primarily affects persons aged 65 and older. It leads to a decline in motor function as a result of the buildup of abnormal protein deposits called Lewy bodies in the brain. Existing therapies exhibit restricted effectiveness and undesirable side effects. The objective was to discover potent medications that have demonstrated effectiveness in treating PD by employing computational methods. This work employed a comprehensive approach to evaluate 70 pyrimidine derivatives for their potential in treating PD. The evaluation involved the use of QSAR modelling, virtual screening, molecular docking, MD simulation, ADMET analysis, and antagonist inhibitor creation. Six compounds passed all the evaluation, while for MD simulation, carried out between the compound with best docking score and the reference drug, compound 57 was discovered to possess more stability compared to theophylline which is the reference drug, and it functions as a primary inhibitor of the adenosine A2A receptor. Additionally, the study determined that compound 57 satisfied the Rule of Five (Ro5) standards and exhibited robust physicochemical characteristics. The compound exhibited moderate to low levels of hERG inhibition. The conducted investigations highlighted promising outcomes of natural compounds that can orient towards the rational development of effective Parkinson's disease inhibitors.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40203-025-00305-9.

Special Issue "Advances in Drug Discovery and Synthesis"

In modern medicinal chemistry, drug discovery is a long, difficult, highly expensive and highly risky process for the identification of new drug compounds [...].

Unlocking Patient Resistance to AI in Healthcare: A Psychological Exploration

Artificial intelligence (AI) has transformed healthcare, yet patients' acceptance of AI-driven medical services remains constrained. Despite its significant potential, patients exhibit reluctance towards this technology. A notable lack of comprehensive research exists that examines the variables driving patients' resistance to AI. This study explores the variables influencing patients' resistance to adopt AI technology in healthcare by applying an extended Ram and Sheth Model. More specifically, this research examines the roles of the need for personal contact (NPC), perceived technological dependence (PTD), and general skepticism toward AI (GSAI) in shaping patient resistance to AI integration. For this reason, a sequential mixed-method approach was employed, beginning with semi-structured interviews to identify adaptable factors in healthcare. It then followed with a survey to validate the qualitative findings through Structural Equation Modeling (SEM) via AMOS (version 24). The findings confirm that NPC, PTD, and GSAI significantly contribute to patient resistance to AI in healthcare. Precisely, patients who prefer personal interaction, feel dependent on AI, or are skeptical of AI's promises are more likely to resist its adoption. The findings highlight the psychological factors driving patient reluctance toward AI in healthcare, offering valuable insights for healthcare administrators. Strategies to balance AI's efficiency with human interaction, mitigate technological dependence, and foster trust are recommended for successful implementation of AI. This research adds to the theoretical understanding of Innovation Resistance Theory, providing both conceptual insights and practical implications for the effective incorporation of AI in healthcare.

GDMol: Generative Double-Masking Self-Supervised Learning for Molecular Property Prediction

BACKGROUND

Effective molecular feature representation is crucial for drug property prediction. Recent years have seen increased attention on graph neural networks (GNNs) that are pre-trained using self-supervised learning techniques, aiming to overcome the scarcity of labeled data in molecular property prediction. Traditional GNNs in self-supervised molecular property prediction typically perform a single masking operation on the nodes and edges of the input molecular graph, masking only local information and insufficient for thorough self-supervised training.

METHOD

Hence, we propose a model for molecular property prediction based on generative double-masking self-supervised learning, termed as GDMol. This integrates generative learning into the self-supervised learning framework for latent representation, and applies a second round of masking to these latent representations, enabling the model to better capture global information and semantic knowledge of the molecules for a richer, more informative representation, thereby achieving more accurate and robust molecular property prediction.

RESULTS

Our experiments on 5 datasets demonstrated superior performance of GDMol in predicting molecular properties across different domains. Moreover, we used the masking operation to traverse through the gradient changes of each node, the magnitude and sign of which reflect the positive and negative contribution respectively of the local structure in the molecule to the prediction outcome. This in-depth interpretative analysis not only enhances the model's interpretability, but also provides more targeted insights and direction for optimizing drug molecules.

CONCLUSIONS

In summary, this research offers novel insights on improving molecular property prediction tasks, and paves the way for further research on the application of generative learning and self-supervised learning in the field of chemistry.

Prediction of bladder cancer prognosis and immune microenvironment assessment using machine learning and deep learning models

Bladder cancer (BCa) is a heterogeneous malignancy characterized by distinct immune subtypes, primarily due to differences in tumor-infiltrating immune cells and their functional characteristics. Therefore, understanding the tumor immune microenvironment (TIME) landscape in BCa is crucial for prognostic prediction and guiding precision therapy. In this study, we integrated 10 machine learning algorithms to develop an immune-related machine learning signature (IRMLS) and subsequently created a deep learning model to detect the IRMLS subtype based on pathological images. The IRMLS proved to be an independent prognostic factor for overall survival (OS) and demonstrated robust and stable performance (p < 0.01). The high-risk group exhibited an immune-inflamed phenotype, associated with poorer prognosis and higher levels of immune cell infiltration. We further investigated the cancer immune cycle and mutation landscape within the IRMLS model, confirming that the high-risk group is more sensitive to immune checkpoint immunotherapy (ICI) and adjuvant chemotherapy with cisplatin (p = 2.8e-10), docetaxel (p = 8.8e-13), etoposide (p = 1.8e-07), and paclitaxel (p = 6.2e-13). In conclusion, we identified and validated a machine learning-based molecular characteristic, IRMLS, which reflects various aspects of the BCa biological process and offers new insights into personalized precision therapy for BCa patients.

Artificial intelligence in antidiabetic drug discovery: The advances in QSAR and the prediction of α-glucosidase inhibitors

Artificial Intelligence is transforming drug discovery, particularly in the hit identification phase of therapeutic compounds. One tool that has been instrumental in this transformation is Quantitative Structure-Activity Relationship (QSAR) analysis. This computer-aided drug design tool uses machine learning to predict the biological activity of new compounds based on the numerical representation of chemical structures against various biological targets. With diabetes mellitus becoming a significant health challenge in recent times, there is intense research interest in modulating antidiabetic drug targets. α-Glucosidase is an antidiabetic target that has gained attention due to its ability to suppress postprandial hyperglycaemia, a key contributor to diabetic complications. This review explored a detailed approach to developing QSAR models, focusing on strategies for generating input variables (molecular descriptors) and computational approaches ranging from classical machine learning algorithms to modern deep learning algorithms. We also highlighted studies that have used these approaches to develop predictive models for α-glucosidase inhibitors to modulate this critical antidiabetic drug target.

Machine learning and artificial intelligence: Enabling the clinical translation of atomic force microscopy-based biomarkers for cancer diagnosis

The influence of biomechanics on cell function has become increasingly defined over recent years. Biomechanical changes are known to affect oncogenesis; however, these effects are not yet fully understood. Atomic force microscopy (AFM) is the gold standard method for measuring tissue mechanics on the micro- or nano-scale. Due to its complexity, however, AFM has yet to become integrated in routine clinical diagnosis. Artificial intelligence (AI) and machine learning (ML) have the potential to make AFM more accessible, principally through automation of analysis. In this review, AFM and its use for the assessment of cell and tissue mechanics in cancer is described. Research relating to the application of artificial intelligence and machine learning in the analysis of AFM topography and force spectroscopy of cancer tissue and cells are reviewed. The application of machine learning and artificial intelligence to AFM has the potential to enable the widespread use of nanoscale morphologic and biomechanical features as diagnostic and prognostic biomarkers in cancer treatment.

tcrBLOSUM: an amino acid substitution matrix for sensitive alignment of distant epitope-specific TCRs

Deciphering the specificity of T-cell receptor (TCR) repertoires is crucial for monitoring adaptive immune responses and developing targeted immunotherapies and vaccines. To elucidate the specificity of previously unseen TCRs, many methods employ the BLOSUM62 matrix to find TCRs with similar amino acid (AA) sequences. However, while BLOSUM62 reflects the AA substitutions within conserved regions of proteins with similar functions, the remarkable diversity of TCRs means that both TCRs with similar and dissimilar sequences can bind the same epitope. Therefore, reliance on BLOSUM62 may bias detection towards epitope-specific TCRs with similar biochemical properties, overlooking those with more diverse AA compositions. In this study, we introduce tcrBLOSUMa and tcrBLOSUMb, specialized AA substitution matrices for CDR3 alpha and CDR3 beta TCR chains, respectively. The matrices reflect AA frequencies and variations occurring within TCRs that bind the same epitope, revealing that both CDR3 alpha and CDR3 beta display tolerance to a wide range of AA substitutions and differ noticeably from the standard BLOSUM62. By accurately aligning distant TCRs employing tcrBLOSUMb, we were able to improve clustering performance and capture a large number of epitope-specific TCRs with diverse AA compositions and physicochemical profiles overlooked by BLOSUM62. Utilizing both the general BLOSUM62 and specialized tcrBLOSUM matrices in existing computational tools will broaden the range of TCRs that can be associated with their cognate epitopes, thereby enhancing TCR repertoire analysis.

In Silico Study of the Potential Inhibitory Effects on Escherichia coli DNA Gyrase of Some Hypothetical Fluoroquinolone-Tetracycline Hybrids

BACKGROUND/OBJECTIVES

Despite the discovery of antibiotics, bacterial infections persist globally, exacerbated by rising antimicrobial resistance that results in millions of cases, increased healthcare costs, and more extended hospital stays. The urgent need for new antibacterial drugs continues as resistance evolves. Fluoroquinolones and tetracyclines are versatile antibiotics that are effective against various bacterial infections. A hybrid antibiotic combines two or more molecules to enhance antimicrobial effectiveness and combat resistance better than monotherapy. Fluoroquinolones are ideal candidates for hybridization due to their potent bactericidal effects, ease of synthesis, and ability to form combinations with other molecules.

METHODS

This study explored the mechanisms of action for 40 hypothetical fluoroquinolone-tetracycline hybrids, all of which could be obtained using a simple, eco-friendly synthesis method. Their interaction with Escherichia coli DNA Gyrase and similarity to albicidin were evaluated using the FORECASTER platform.

RESULTS

Hybrids such as Do-Ba, Mi-Fi, and Te-Ba closely resembled albicidin in physicochemical properties and FITTED Scores, while Te-De surpassed it with a better score. Similar to fluoroquinolones, these hybrids likely inhibit DNA synthesis by binding to enzyme-DNA complexes.

CONCLUSIONS

These hybrids could offer broad-spectrum activity and help mitigate bacterial resistance, though further in vitro and in vivo studies are needed to validate their potential.

DeepLigType: Predicting Ligand Types of ProteinLigand Binding Sites Using a Deep Learning Model

The analysis of protein-ligand binding sites plays a crucial role in the initial stages of drug discovery. Accurately predicting the ligand types that are likely to bind to protein-ligand binding sites enables more informed decision making in drug design. Our study, DeepLigType, determines protein-ligand binding sites using Fpocket and then predicts the ligand type of these pockets with the deep learning model, Convolutional Block Attention Module (CBAM) with ResNet. CBAM-ResNet has been trained to accurately predict five distinct ligand types. We classified protein-ligand binding sites into five different categories according to the type of response ligands cause when they bind to their target proteins, which are antagonist, agonist, activator, inhibitor, and others. We created a novel dataset, referred to as LigType5, from the widely recognized PDBbind and scPDB dataset for training and testing our model. While the literature mostly focuses on the specificity and characteristic analysis of protein binding sites by experimental (laboratory-based) methods, we propose a computational method with the DeepLigType architecture. DeepLigType demonstrated an accuracy of 74.30% and an AUC of 0.83 in ligand type prediction on a novel test dataset using the CBAM-ResNet deep learning model.

A comprehensive survey of drug-target interaction analysis in allopathy and siddha medicine

Effective drug delivery is the cornerstone of modern healthcare, ensuring therapeutic compounds reach their intended targets efficiently. This paper explores the potential of personalized and holistic healthcare, driven by the synergy between traditional and allopathic medicine systems, with a specific focus on the vast reservoir of medicinal compounds found in plants rooted in the historical legacy of traditional medicine. Motivated by the desire to unlock the therapeutic potential of medicinal plants and bridge the gap between traditional and allopathic medicine, this survey delves into in-silico computational approaches for studying Drug-Target Interactions (DTI) within the contexts of allopathy and siddha medicine. The contributions of this survey are multifaceted: it offers a comprehensive overview of in-silico methods for DTI analysis in both systems, identifies common challenges in DTI studies, provides insights into future directions to advance DTI analysis, and includes a comparative analysis of DTI in allopathy and siddha medicine. The findings of this survey highlight the pivotal role of in-silico computational approaches in advancing drug research and development in both allopathy and siddha medicine, emphasizing the importance of integrating these methods to drive the future of personalized healthcare.

In vitro neurotoxicity testing: lessons from chemotherapy-induced peripheral neurotoxicity

INTRODUCTION

Chemotherapy induced peripheral neurotoxicity (CIPN) is a long-lasting, or even permanent, late toxicity caused by largely used anticancer drugs. CIPN affects a growing population of cancer survivors and diminishes their quality of life since there is no curative/preventive treatment. Among several reasons for this unmet clinical need, there is an incomplete knowledge on mechanisms leading to CIPN. Therefore, bench side research is still greatly needed: in vitro studies are pivotal to both evaluate neurotoxicity mechanisms and potential neuroprotection strategies.

AREAS COVERED

Advantages and disadvantages of in vitro approaches are addressed with respect to their applicability to the CIPN field. Different cell cultures and techniques to assess neurotoxicity/neuroprotection are described. PubMed search-string: (chemotherapy-induced) AND (((neuropathy) OR neurotoxicity) OR neuropathic pain) AND (in vitro) AND (((((model) OR SH-SY5Y) OR PC12) OR iPSC) OR DRG neurons); (chemotherapy-induced) AND (((neuropathy) OR neurotoxicity) OR neuropathic pain) AND (model) AND (((neurite elongation) OR cell viability) OR morphology). No articles published before 1990 were selected.

EXPERT OPINION

CIPN is an ideal experimental setting to test axonal damage and, in general, peripheral nervous system mechanisms of disease and neuroprotection. Therefore, starting from robust preclinical data in this field, potentially, relevant biological rationale can be transferred to other human spontaneous diseases of the peripheral nervous system.

Artificial Intelligence (AI) Applications in Drug Discovery and Drug Delivery: Revolutionizing Personalized Medicine

Artificial intelligence (AI) encompasses a broad spectrum of techniques that have been utilized by pharmaceutical companies for decades, including machine learning, deep learning, and other advanced computational methods. These innovations have unlocked unprecedented opportunities for the acceleration of drug discovery and delivery, the optimization of treatment regimens, and the improvement of patient outcomes. AI is swiftly transforming the pharmaceutical industry, revolutionizing everything from drug development and discovery to personalized medicine, including target identification and validation, selection of excipients, prediction of the synthetic route, supply chain optimization, monitoring during continuous manufacturing processes, or predictive maintenance, among others. While the integration of AI promises to enhance efficiency, reduce costs, and improve both medicines and patient health, it also raises important questions from a regulatory point of view. In this review article, we will present a comprehensive overview of AI's applications in the pharmaceutical industry, covering areas such as drug discovery, target optimization, personalized medicine, drug safety, and more. By analyzing current research trends and case studies, we aim to shed light on AI's transformative impact on the pharmaceutical industry and its broader implications for healthcare.

Machine Learning-Assisted Mixed Quantum-Classical Dynamics without Explicit Nonadiabatic Coupling: Application to the Photodissociation of Peroxynitric Acid

We have devised a hybrid quantum-classical scheme utilizing machine-learned potential energy surfaces (PES), which circumvents the need for explicit computation of nonadiabatic coupling elements. The quantities necessary to account for the nonadiabatic effects are directly obtained from the PESs. The simulation of dynamics is based on the fewest-switches surface-hopping method. We applied this scheme to model the photodissociation of both N-O and O-O bonds in a conformer of peroxynitric acid (HO2NO2). Adiabatic PES data for the six lowest states of this molecule were computed at the CASSCF level for various nuclear configurations. These served as the training data for the machine-learning models for the PESs. The dynamics simulation was initiated on the lowest optically bright singlet excited state (S4) and propagated along the two Jacobi coordinates J → 1 and J → 2 while accounting for the nonadiabatic effects through transitions between PESs. Our analysis revealed that there is a very high chance of dissociation of the N-O bond leading to the HO2 and NO2 fragments.

Brain-on-a-chip: an emerging platform for studying the nanotechnology-biology interface for neurodegenerative disorders

Neurological disorders have for a long time been a global challenge dismissed by drug companies, especially due to the low efficiency of most therapeutic compounds to cross the brain capillary wall, that forms the blood-brain barrier (BBB) and reach the brain. This has boosted an incessant search for novel carriers and methodologies to drive these compounds throughout the BBB. However, it remains a challenge to artificially mimic the physiology and function of the human BBB, allowing a reliable, reproducible and throughput screening of these rapidly growing technologies and nanoformulations (NFs). To surpass these challenges, brain-on-a-chip (BoC) - advanced microphysiological platforms that emulate key features of the brain composition and functionality, with the potential to emulate pathophysiological signatures of neurological disorders, are emerging as a microfluidic tool to screen new brain-targeting drugs, investigate neuropathogenesis and reach personalized medicine. In this review, the advance of BoC as a bioengineered screening tool of new brain-targeting drugs and NFs, enabling to decipher the intricate nanotechnology-biology interface is discussed. Firstly, the main challenges to model the brain are outlined, then, examples of BoC platforms to recapitulate the neurodegenerative diseases and screen NFs are summarized, emphasizing the current most promising nanotechnological-based drug delivery strategies and lastly, the integration of high-throughput screening biosensing systems as possible cutting-edge technologies for an end-use perspective is discussed as future perspective.

Machine learning assisted sorting of active microswimmers

Active matter systems, being in a non-equilibrium state, exhibit complex behaviors, such as self-organization, giving rise to emergent phenomena. There are many examples of active particles with biological origins, including bacteria and spermatozoa, or with artificial origins, such as self-propelled swimmers and Janus particles. The ability to manipulate active particles is vital for their effective application, e.g., separating motile spermatozoa from nonmotile and dead ones, to increase fertilization chance. In this study, we proposed a mechanism-an apparatus-to sort and demix active particles based on their motility values (Péclet number). Initially, using Brownian simulations, we demonstrated the feasibility of sorting self-propelled particles. Following this, we employed machine learning methods, supplemented with data from comprehensive simulations that we conducted for this study, to model the complex behavior of active particles. This enabled us to sort them based on their Péclet number. Finally, we evaluated the performance of the developed models and showed their effectiveness in demixing and sorting the active particles. Our findings can find applications in various fields, including physics, biology, and biomedical science, where the sorting and manipulation of active particles play a pivotal role.

Comparison of the efficacy and adverse effects of oral ferrous succinate tablets and intravenous iron sucrose: a retrospective study

OBJECTIVE

To analyse the clinical efficacy and adverse drug reactions (ADRs) of iron preparations.

METHODS

A total of 374 patients with iron deficiency anaemia admitted to our hospital between 1 January and 31 December 2020 were included in this study. They were divided into 2 groups based on their medication regimens: Group A (n = 187) took oral ferrous succinate tablets, and Group B (n = 187) received intravenous iron sucrose. The remission of major symptoms, laboratory test results, ADRs and other related data were collected after 4 weeks of treatment.

RESULTS

Compared with the pre-treatment baseline, haemoglobin (Hb), serum iron (SI), serum ferritin (SF) and the mean corpuscular volume (MCV) increased in both groups at 4 weeks of treatment (P < 0.05). After treatment, Group A had lower levels of Hb (108.41 ± 8.39 vs. 122.31 ± 6.04 g/L, t = 6.293, P < 0.001), SI (9.72 ± 4.24 vs. 15.62 ± 5.41 µmol/L, t = 5.482, P < 0.001) and SF (27.1 ± 10.82 vs. 39.82 ± 10.44 ug/L, t = 6.793, P < 0.001) compared with Group B. In contrast, there was no significant difference in the post-treatment level of MCV (P > 0.05). The overall response rate significantly differed between the 2 groups (78.61% vs. 90.91%, χ2 = 10.949, P < 0.001). The incidence of ADRs of both groups were similar, and the difference was not statistically significant (χ2 = 0.035, P = 0.851).

CONCLUSION

Iron sucrose demonstrates favourable efficacy and safety in treating iron deficiency anaemia.

A Chronicle Review of In-Silico Approaches for Discovering Novel Antimicrobial Agents to Combat Antimicrobial Resistance

Antimicrobial resistance (AMR) poses a foremost threat to global health, necessitating innovative strategies for discovering antimicrobial agents. This review explores the role and recent advances of in-silico techniques in identifying novel antimicrobial agents and combating AMR giving few briefings of recent case studies of AMR. In-silico techniques, such as homology modeling, virtual screening, molecular docking, pharmacophore modeling, molecular dynamics simulation, density functional theory, integrated machine learning, and artificial intelligence, are systematically reviewed for their utility in discovering antimicrobial agents. These computational methods enable the rapid screening of large compound libraries, prediction of drug-target interactions, and optimization of drug candidates. The review discusses integrating in-silico approaches with traditional experimental methods and highlights their potential to accelerate the discovery of new antimicrobial agents. Furthermore, it emphasizes the significance of interdisciplinary collaboration and data-sharing initiatives in advancing antimicrobial research. Through a comprehensive discussion of the latest developments in in-silico techniques, this review provides valuable insights into the future of antimicrobial research and the fight against AMR.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s12088-024-01355-x.

Democratizing cheminformatics: interpretable chemical grouping using an automated KNIME workflow

With the increased availability of chemical data in public databases, innovative techniques and algorithms have emerged for the analysis, exploration, visualization, and extraction of information from these data. One such technique is chemical grouping, where chemicals with common characteristics are categorized into distinct groups based on physicochemical properties, use, biological activity, or a combination. However, existing tools for chemical grouping often require specialized programming skills or the use of commercial software packages. To address these challenges, we developed a user-friendly chemical grouping workflow implemented in KNIME, a free, open-source, low/no-code, data analytics platform. The workflow serves as an all-encompassing tool, expertly incorporating a range of processes such as molecular descriptor calculation, feature selection, dimensionality reduction, hyperparameter search, and supervised and unsupervised machine learning methods, enabling effective chemical grouping and visualization of results. Furthermore, we implemented tools for interpretation, identifying key molecular descriptors for the chemical groups, and using natural language summaries to clarify the rationale behind these groupings. The workflow was designed to run seamlessly in both the KNIME local desktop version and KNIME Server WebPortal as a web application. It incorporates interactive interfaces and guides to assist users in a step-by-step manner. We demonstrate the utility of this workflow through a case study using an eye irritation and corrosion dataset.Scientific contributionsThis work presents a novel, comprehensive chemical grouping workflow in KNIME, enhancing accessibility by integrating a user-friendly graphical interface that eliminates the need for extensive programming skills. This workflow uniquely combines several features such as automated molecular descriptor calculation, feature selection, dimensionality reduction, and machine learning algorithms (both supervised and unsupervised), with hyperparameter optimization to refine chemical grouping accuracy. Moreover, we have introduced an innovative interpretative step and natural language summaries to elucidate the underlying reasons for chemical groupings, significantly advancing the usability of the tool and interpretability of the results.

Machine-learning-based models assist the prediction of pulmonary embolism in autoimmune diseases: A retrospective, multicenter study

BACKGROUND

Pulmonary embolism (PE) is a severe and acute cardiovascular syndrome with high mortality among patients with autoimmune inflammatory rheumatic diseases (AIIRDs). Accurate prediction and timely intervention play a pivotal role in enhancing survival rates. However, there is a notable scarcity of practical early prediction and risk assessment systems of PE in patients with AIIRD.

METHODS

In the training cohort, 60 AIIRD with PE cases and 180 age-, gender-, and disease-matched AIIRD non-PE cases were identified from 7254 AIIRD cases in Tongji Hospital from 2014 to 2022. Univariable logistic regression (LR) and least absolute shrinkage and selection operator (LASSO) were used to select the clinical features for further training with machine learning (ML) methods, including random forest (RF), support vector machines (SVM), neural network (NN), logistic regression (LR), gradient boosted decision tree (GBDT), classification and regression trees (CART), and C5.0 models. The performances of these models were subsequently validated using a multicenter validation cohort.

RESULTS

In the training cohort, 24 and 13 clinical features were selected by univariable LR and LASSO strategies, respectively. The five ML models (RF, SVM, NN, LR, and GBDT) showed promising performances, with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.962-1.000 in the training cohort and 0.969-0.999 in the validation cohort. CART and C5.0 models achieved AUCs of 0.850 and 0.932, respectively, in the training cohort. Using D-dimer as a pre-screening index, the refined C5.0 model achieved an AUC exceeding 0.948 in the training cohort and an AUC above 0.925 in the validation cohort. These results markedly outperformed the use of D-dimer levels alone.

CONCLUSION

ML-based models are proven to be precise for predicting the onset of PE in patients with AIIRD exhibiting clinical suspicion of PE.

TRIAL REGISTRATION

Chictr.org.cn : ChiCTR2200059599.

scDrug+: predicting drug-responses using single-cell transcriptomics and molecular structure

Predicting drug responses based on individual transcriptomic profiles holds promise for refining prognosis and advancing precision medicine. Although many studies have endeavored to predict the responses of known drugs to novel transcriptomic profiles, research into predicting responses for newly discovered drugs remains sparse. In this study, we introduce scDrug+, a comprehensive pipeline that seamlessly integrates single-cell analysis with drug-response prediction. Importantly, scDrug+ is equipped to predict the response of new drugs by analyzing their molecular structures. The open-source tool is available as a Docker container, ensuring ease of deployment and reproducibility. It can be accessed at https://github.com/ailabstw/scDrugplus.

Explainable artificial intelligence-assisted virtual screening and bioinformatics approaches for effective bioactivity prediction of phenolic cyclooxygenase-2 (COX-2) inhibitors using PubChem molecular fingerprints

Cyclooxygenase-2 (COX-2) inhibitors are nonsteroidal anti-inflammatory drugs that treat inflammation, pain and fever. This study determined the interaction mechanisms of COX-2 inhibitors and the molecular properties needed to design new drug candidates. Using machine learning and explainable AI methods, the inhibition activity of 1488 molecules was modelled, and essential properties were identified. These properties included aromatic rings, nitrogen-containing functional groups and aliphatic hydrocarbons. They affected the water solubility, hydrophobicity and binding affinity of COX-2 inhibitors. The binding mode, stability and ADME properties of 16 ligands bound to the Cyclooxygenase active site of COX-2 were investigated by molecular docking, molecular dynamics simulation and MM-GBSA analysis. The results showed that ligand 339,222 was the most stable and effective COX-2 inhibitor. It inhibited prostaglandin synthesis by disrupting the protein conformation of COX-2. It had good ADME properties and high clinical potential. This study demonstrated the potential of machine learning and bioinformatics methods in discovering COX-2 inhibitors.

Artificial intelligence for small molecule anticancer drug discovery

INTRODUCTION

The transition from conventional cytotoxic chemotherapy to targeted cancer therapy with small-molecule anticancer drugs has enhanced treatment outcomes. This approach, which now dominates cancer treatment, has its advantages. Despite the regulatory approval of several targeted molecules for clinical use, challenges such as low response rates and drug resistance still persist. Conventional drug discovery methods are costly and time-consuming, necessitating more efficient approaches. The rise of artificial intelligence (AI) and access to large-scale datasets have revolutionized the field of small-molecule cancer drug discovery. Machine learning (ML), particularly deep learning (DL) techniques, enables the rapid identification and development of novel anticancer agents by analyzing vast amounts of genomic, proteomic, and imaging data to uncover hidden patterns and relationships.

AREA COVERED

In this review, the authors explore the important landmarks in the history of AI-driven drug discovery. They also highlight various applications in small-molecule cancer drug discovery, outline the challenges faced, and provide insights for future research.

EXPERT OPINION

The advent of big data has allowed AI to penetrate and enable innovations in almost every stage of medicine discovery, transforming the landscape of oncology research through the development of state-of-the-art algorithms and models. Despite challenges in data quality, model interpretability, and technical limitations, advancements promise breakthroughs in personalized and precision oncology, revolutionizing future cancer management.

In Silico Design and In Vitro Assessment of Bicyclic Trifluoromethylated Pyrroles as New Antibacterial and Antifungal Agents

QSAR studies on the number of compounds tested as S. aureus inhibitors were performed using an interactive Online Chemical Database and Modeling Environment (OCHEM) web platform. The predictive ability of the developed consensus QSAR model was q2=0.79±0.02. The consensus prediction for the external evaluation set afforded high predictive power (q2=0.82±0.03). The models were applied to screen a virtual chemical library with anti-S. aureus activity. Six promising new bicyclic trifluoromethylated pyrroles were identified, synthesized and evaluated in vitro against S. aureus, E. coli, and A. baumannii for their antibacterial activity and against C. albicans, C. krusei and C. glabrata for their antifungal activity. The synthesized compounds were characterized by 1H, 19F, and 13C NMR and elemental analysis. The antimicrobial activity assessment indicated that trifluoromethylated pyrroles 9 and 11 demonstrated the greatest antibacterial and antifungal effects against all the tested pathogens, especially against multidrug-resistant strains. The acute toxicity of the compounds to Daphnia magna ranged from 1.21 to 33.39 mg/L (moderately and slightly toxic). Based on the docking results, it can be suggested that the antibacterial and antifungal effects of the compounds can be explained by the inhibition of bacterial wall component synthesis.

Advances in artificial intelligence for drug delivery and development: A comprehensive review

Artificial intelligence (AI) has emerged as a powerful tool to revolutionize the healthcare sector, including drug delivery and development. This review explores the current and future applications of AI in the pharmaceutical industry, focusing on drug delivery and development. It covers various aspects such as smart drug delivery networks, sensors, drug repurposing, statistical modeling, and simulation of biotechnological and biological systems. The integration of AI with nanotechnologies and nanomedicines is also examined. AI offers significant advancements in drug discovery by efficiently identifying compounds, validating drug targets, streamlining drug structures, and prioritizing response templates. Techniques like data mining, multitask learning, and high-throughput screening contribute to better drug discovery and development innovations. The review discusses AI applications in drug formulation and delivery, clinical trials, drug safety, and pharmacovigilance. It addresses regulatory considerations and challenges associated with AI in pharmaceuticals, including privacy, data security, and interpretability of AI models. The review concludes with future perspectives, highlighting emerging trends, addressing limitations and biases in AI models, and emphasizing the importance of collaboration and knowledge sharing. It provides a comprehensive overview of AI's potential to transform the pharmaceutical industry and improve patient care while identifying further research and development areas.

Modelling the nicotine pharmacokinetic profile for e-cigarettes using real time monitoring of consumers' physiological measurements and mouth level exposure

Pharmacokinetic (PK) studies can provide essential information on abuse liability of nicotine and tobacco products but are intrusive and must be conducted in a clinical environment. The objective of the study was to explore whether changes in plasma nicotine levels following use of an e-cigarette can be predicted from real time monitoring of physiological parameters and mouth level exposure (MLE) to nicotine before, during, and after e-cigarette vaping, using wearable devices. Such an approach would allow an -effective pre-screening process, reducing the number of clinical studies, reducing the number of products to be tested and the number of blood draws required in a clinical PK study Establishing such a prediction model might facilitate the longitudinal collection of data on product use and nicotine expression among consumers using nicotine products in their normal environments, thereby reducing the need for intrusive clinical studies while generating PK data related to product use in the real world.An exploratory machine learning model was developed to predict changes in plasma nicotine levels following the use of an e-cigarette; from real time monitoring of physiological parameters and MLE to nicotine before, during, and after e-cigarette vaping. This preliminary study identified key parameters, such as heart rate (HR), heart rate variability (HRV), and physiological stress (PS) that may act as predictors for an individual's plasma nicotine response (PK curve). Relative to baseline measurements (per participant), HR showed a significant increase for nicotine containing e-liquids and was consistent across sessions (intra-participant). Imputing missing values and training the model on all data resulted in 57% improvement from the original'learning' data and achieved a median validation R2 of 0.70.The study is in its exploratory phase, with limitations including a small and non-diverse sample size and reliance on data from a single e-cigarette product. These findings necessitate further research for validation and to enhance the model's generalisability and applicability in real-world settings. This study serves as a foundational step towards developing non-intrusive PK models for nicotine product use.

Integrated ML-Based Strategy Identifies Drug Repurposing for Idiopathic Pulmonary Fibrosis

Idiopathic pulmonary fibrosis (IPF) affects an estimated global population of around 3 million individuals. IPF is a medical condition with an unknown cause characterized by the formation of scar tissue in the lungs, leading to progressive respiratory disease. Currently, there are only two FDA-approved small molecule drugs specifically for the treatment of IPF and this has created a demand for the rapid development of drugs for IPF treatment. Moreover, denovo drug development is time and cost-intensive with less than a 10% success rate. Drug repurposing currently is the most feasible option for rapidly making the drugs to market for a rare and sporadic disease. Normally, the repurposing of drugs begins with a screening of FDA-approved drugs using computational tools, which results in a low hit rate. Here, an integrated machine learning-based drug repurposing strategy is developed to significantly reduce the false positive outcomes by introducing the predock machine-learning-based predictions followed by literature and GSEA-assisted validation and drug pathway prediction. The developed strategy is deployed to 1480 FDA-approved drugs and to drugs currently in a clinical trial for IPF to screen them against "TGFB1", "TGFB2", "PDGFR-a", "SMAD-2/3", "FGF-2", and more proteins resulting in 247 total and 27 potentially repurposable drugs. The literature and GSEA validation suggested that 72 of 247 (29.14%) drugs have been tried for IPF, 13 of 247 (5.2%) drugs have already been used for lung fibrosis, and 20 of 247 (8%) drugs have been tested for other fibrotic conditions such as cystic fibrosis and renal fibrosis. Pathway prediction of the remaining 142 drugs was carried out resulting in 118 distinct pathways. Furthermore, the analysis revealed that 29 of 118 pathways were directly or indirectly involved in IPF and 11 of 29 pathways were directly involved. Moreover, 15 potential drug combinations are suggested for showing a strong synergistic effect in IPF. The drug repurposing strategy reported here will be useful for rapidly developing drugs for treating IPF and other related conditions.

Predicting blood-brain barrier permeability of molecules with a large language model and machine learning

Predicting the blood-brain barrier (BBB) permeability of small-molecule compounds using a novel artificial intelligence platform is necessary for drug discovery. Machine learning and a large language model on artificial intelligence (AI) tools improve the accuracy and shorten the time for new drug development. The primary goal of this research is to develop artificial intelligence (AI) computing models and novel deep learning architectures capable of predicting whether molecules can permeate the human blood-brain barrier (BBB). The in silico (computational) and in vitro (experimental) results were validated by the Natural Products Research Laboratories (NPRL) at China Medical University Hospital (CMUH). The transformer-based MegaMolBART was used as the simplified molecular input line entry system (SMILES) encoder with an XGBoost classifier as an in silico method to check if a molecule could cross through the BBB. We used Morgan or Circular fingerprints to apply the Morgan algorithm to a set of atomic invariants as a baseline encoder also with an XGBoost classifier to compare the results. BBB permeability was assessed in vitro using three-dimensional (3D) human BBB spheroids (human brain microvascular endothelial cells, brain vascular pericytes, and astrocytes). Using multiple BBB databases, the results of the final in silico transformer and XGBoost model achieved an area under the receiver operating characteristic curve of 0.88 on the held-out test dataset. Temozolomide (TMZ) and 21 randomly selected BBB permeable compounds (Pred scores = 1, indicating BBB-permeable) from the NPRL penetrated human BBB spheroid cells. No evidence suggests that ferulic acid or five BBB-impermeable compounds (Pred scores < 1.29423E-05, which designate compounds that pass through the human BBB) can pass through the spheroid cells of the BBB. Our validation of in vitro experiments indicated that the in silico prediction of small-molecule permeation in the BBB model is accurate. Transformer-based models like MegaMolBART, leveraging the SMILES representations of molecules, show great promise for applications in new drug discovery. These models have the potential to accelerate the development of novel targeted treatments for disorders of the central nervous system.

Explainable artificial intelligence in the design of selective carbonic anhydrase I-II inhibitors via molecular fingerprinting

Inhibiting the enzymes carbonic anhydrase I (CA I) and carbonic anhydrase II (CA II) presents a potential avenue for addressing nervous system ailments such as glaucoma and Alzheimer's disease. Our study explored harnessing explainable artificial intelligence (XAI) to unveil the molecular traits inherent in CA I and CA II inhibitors. The PubChem molecular fingerprints of these inhibitors, sourced from the ChEMBL database, were subjected to detailed XAI analysis. The study encompassed training 10 regression models using IC50 values, and their efficacy was gauged using metrics including R2, RMSE, and time taken. The Decision Tree Regressor algorithm emerged as the optimal performer (R2: 0.93, RMSE: 0.43, time-taken: 0.07). Furthermore, the PFI method unveiled key molecular features for CA I inhibitors, notably PubChemFP432 (C(O)N) and PubChemFP6978 (C(O)O). The SHAP analysis highlighted the significance of attributes like PubChemFP539 (C(O)NCC), PubChemFP601 (C(O)OCC), and PubChemFP432 (C(O)N) in CA I inhibitiotable n. Likewise, features for CA II inhibitors encompassed PubChemFP528(C(O)OCCN), PubChemFP791 (C(O)OCCC), PubChemFP696 (C(O)OCCCC), PubChemFP335 (C(O)NCCN), PubChemFP580 (C(O)NCCCN), and PubChemFP180 (C(O)NCCC), identified through SHAP analysis. The sulfonamide group (S), aromatic ring (A), and hydrogen bonding group (H) exert a substantial impact on CA I and CA II enzyme activities and IC50 values through the XAI approach. These insights into the CA I and CA II inhibitors are poised to guide future drug discovery efforts, serving as a beacon for innovative therapeutic interventions.