Machine Learning Study of Metabolic Networks vs ChEMBL Data of Antibacterial Compounds

First report on Quantitative Structure-Toxicity Relationship modeling approaches for the prediction of acute toxicity of various organic chemicals against rotifer species

Nowadays, organic chemicals are crucial components in virtually every aspect of daily life, serving as indispensable elements for modern society. The ongoing synthesis of chemicals and the various potential harmful effects on living organisms are prompting regulatory bodies to view computational approaches as vital supplements and alternatives to traditional animal testing in assessing chemical risks. In this study, we have developed, for the first time, Quantitative Structure-Toxicity Relationship (QSTR) models based on Multiple Linear Regression (MLR) and five Machine Learning (ML) algorithms to predict organic chemical toxicity against a rotifer species (Brachionus calyciflorus). The most influential descriptors included in the MLR model are (SM6_B(p), B07[ClCl], B05[ClCl], MaxssCH2, F09[NO], B04[ClCl], and minssO), with positive contributions to the dependent variable (negative decimal logarithm of median lethal concentration at 24 h). The interpretation of the molecular descriptors of the MLR model suggested that substances with high molecular polarizability and lipophilicity (presence of chlorine atoms) positively influence and increase their toxic potency. The analysis of the application domain, conducted using the leverage approach and the standardized residual method, showcased the extensive applicability of each model. In the cross-validation, the best values are presented by Support Vector Regression (SV_R), a value of Q2Loo = 0.754 and RMSEcv = 0.652, which are slightly higher than the results of the other linear and nonlinear techniques used. Furthermore, our research exhibited a high degree of fitness, internal robustness, and external predictive power. These findings suggest that the developed QSTR models are well-suited for the reliable prediction of aquatic toxicity for a wide range of structurally diverse organic chemicals. These models can be valuable for tasks such as screening, prioritizing new compounds, filling data gaps, and mitigating the limitations associated with in vivo and in vitro tests, ultimately contributing to the reduction of the use of dangerous chemicals in the environment.

Prediction of the neurotoxic mechanisms of the pesticide phorate using network toxicology, molecular docking, and molecular dynamics simulation

Phorate is an organophosphate pesticide that may cause neurotoxicity, although the exact mechanisms remain unclear.This study aimed to elucidate the mechanisms of neurotoxicity caused by phorate overexposure using network toxicology, molecular docking, and molecular dynamics simulation.We identified 104 potential targets and 20 core targets associated with phorate-induced neurotoxicity. Key targets, including MMP9, CASP1, and KEAP1, may be involved in neuroactive ligand-receptor interaction signalling, as well as the cAMP and calcium signalling pathways. Furthermore, molecular dynamics simulations were conducted on the KEAP1 and CASP1 protein-ligand complexes, which demonstrated the highest binding stabilities in molecular docking analysis. The binding free energies were calculated to be -27.08 and -22.80 kcal/mol for KEAP1 and CASP1, respectively, indicating that both complexes are thermodynamically stable.The methodology used in this study facilitates the identification and assessment of previously unexplored agrochemical toxicity pathways and molecular mechanisms. These findings suggest a novel approach to controlling pesticide residues and screening drugs.

In Silico Approach for Antibacterial Discovery: PTML Modeling of Virtual Multi-Strain Inhibitors Against Staphylococcus aureus

Background/Objectives: Infectious diseases caused by Staphylococcus aureus (S. aureus) have become alarming health issues worldwide due to the ever-increasing emergence of multidrug resistance. In silico approaches can accelerate the identification and/or design of versatile antibacterial chemicals with the ability to target multiple S. aureus strains with varying degrees of drug resistance. Here, we develop a perturbation theory machine learning model based on a multilayer perceptron neural network (PTML-MLP) for the prediction and design of versatile virtual inhibitors against S. aureus strains. Methods: To develop the PTML-MLP model, chemical and biological data associated with antibacterial activity against S. aureus strains were retrieved from the ChEMBL database. We applied the Box-Jenkins approach to convert the topological indices into multi-label graph-theoretical indices; the latter were used as inputs for the creation of the PTML-MLP model. Results: The PTML-MLP model exhibited accuracy higher than 80% in both training and test sets. The physicochemical and structural interpretation of the PTML-MLP model was performed through the fragment-based topological design (FBTD) approach. Such interpretations permitted the analysis of different molecular fragments with favorable contributions to the multi-strain antibacterial activity and the design of four new drug-like molecules using different fragments as building blocks. The designed molecules were predicted/confirmed by our PTML model as multi-strain inhibitors of diverse S. aureus strains, thus representing promising chemotypes to be considered for future synthesis and biological testing of versatile anti-S. aureus agents. Conclusions: This work envisages promising applications of PTML modeling for early antibacterial drug discovery and related antimicrobial research areas.

Perturbation-theory machine learning for mood disorders: virtual design of dual inhibitors of NET and SERT proteins

Mood disorders affect the daily lives of millions of people worldwide. The search for more efficient therapies for mood disorders remains an active field of research. In silico approaches can accelerate the search for inhibitors against protein targets related to mood disorders. Here, we developed the first model perturbation-theory machine learning model based on a multiplayer perceptron network (PTML-MLP) for the simultaneous prediction and design of virtual dual-target inhibitors against two proteins associated with mood disorders, namely norepinephrine and serotonin transporters (NET and SERT, respectively). The PTML-MLP model had an accuracy of around 80%. From a chemical point of view, the PTML-MLP model could accurately identify both single- and dual-target inhibitors present in the dataset used to build it. Through the application of the fragment-based topological design (FBTD) approach, the molecular descriptors (multi-label graph-based indices) present in the PTML-MLP model were physicochemically and structurally interpreted. Such interpretations enabled (a) the extraction of different molecular fragments with a positive influence on the enhancement of the dual-target activity and (b) the design of four new drug-like molecules by assembling (fusing and/or connecting) several suitable molecular fragments. The designed molecules were predicted by the PTML-MLP model to exhibit dual-target activity against the NET and SERT proteins. These predictions, together with the estimated druglikeness suggest that the designed molecules could be new promising chemotypes to be considered for future synthesis and biological experimentation in the context of treatments for mood disorders.

Recursive dynamics of GspE through machine learning enabled identification of inhibitors

Type II secretion System has been increasingly recognized as a key driver of virulence in many pathogenic bacteria including Achromobacter xylosoxidans. ATPase GspE is the powerhouse of the T2SS. It powers the entire secretion process by binding with ATP and hydrolyzing it. Therefore, targeting it was thought to have a profound effect on the normal functioning of the whole T2SS. A. xylosoxidans is a Gram-negative bacterium that poses a rising concern to immunocompromised people. It is responsible for many opportunistic infections mostly in people with cystic fibrosis. Due to its intrinsic and acquired resistance mechanisms, it is challenging to treat. In this current study, an extensive machine learning-enabled computational investigation was carried out. Drug libraries were screened using machine learning random forest algorithm trained on non-redundant dataset of 8722 antibacterial compounds with reported IC50 values. Active compounds were then further subjected to molecular docking. To unravel the dynamics and better understand the stability of complexes, the top complexes were subjected to MD Simulations followed by various post-simulation analyses including Trajectory analysis, Atom Contacts, SASA, Hydrogen Bond, RDF, binding free energy calculations, PCA, and AFD analysis. Findings from the study unanimously unveiled Asinex-BAS00263070-28551 as the best inhibitor as it instigated the recursive dynamics of the target by making key hydrogen bond interactions with Walker A motif, suggesting it could serve as the promising drug candidate against GspE. Further experimental in-vivo and in-vitro validation is still required to authenticate the therapeutic effects of these drugs.

Advancing Anticancer Drug Discovery: Leveraging Metabolomics and Machine Learning for Mode of Action Prediction by Pattern Recognition

A bottleneck in the development of new anti-cancer drugs is the recognition of their mode of action (MoA). Metabolomics combined with machine learning allowed to predict MoAs of novel anti-proliferative drug candidates, focusing on human prostate cancer cells (PC-3). As proof of concept, 38 drugs are studied with known effects on 16 key processes of cancer metabolism, profiling low molecular weight intermediates of the central carbon and cellular energy metabolism (CCEM) by LC-MS/MS. These metabolic patterns unveiled distinct MoAs, enabling accurate MoA predictions for novel agents by machine learning. The transferability of MoA predictions based on PC-3 cell treatments is validated with two other cancer cell models, i.e., breast cancer and Ewing's sarcoma, and show that correct MoA predictions for alternative cancer cells are possible, but still at some expense of prediction quality. Furthermore, metabolic profiles of treated cells yield insights into intracellular processes, exemplified for drugs inducing different types of mitochondrial dysfunction. Specifically, it is predicted that pentacyclic triterpenes inhibit oxidative phosphorylation and affect phospholipid biosynthesis, as confirmed by respiration parameters, lipidomics, and molecular docking. Using biochemical insights from individual drug treatments, this approach offers new opportunities, including the optimization of combinatorial drug applications.

Perturbation Theory Machine Learning Model for Phenotypic Early Antineoplastic Drug Discovery: Design of Virtual Anti-Lung-Cancer Agents

Lung cancer is the most diagnosed malignant neoplasm worldwide and it is associated with great mortality. Currently, developing antineoplastic agents is a challenging, time-consuming, and costly process. Computational methods can speed up the early discovery of anti-lung-cancer chemicals. Here, we report a perturbation theory machine learning model based on a multilayer perceptron (PTML-MLP) model for phenotypic early antineoplastic drug discovery, enabling the rational design and prediction of new molecules as virtual versatile inhibitors of multiple lung cancer cell lines. The PTML-MLP model achieved an accuracy above 80%. We applied the fragment-based topological design (FBTD) approach to physicochemically and structurally interpret the PTML-MLP model. This enabled the extraction of suitable fragments with a positive influence on anti-lung-cancer activity against the different lung cancer cell lines. By following the aforementioned interpretations, we could assemble several suitable fragments to design four novel molecules, which were predicted by the PTML-MLP model as versatile anti-lung-cancer agents. Such predictions of potent multi-cellular anticancer activity against diverse lung cancer cell lines were rigorously confirmed by a well-established virtual screening tool reported in the literature. The present work envisages new opportunities for the application of PTML models to accelerate early antineoplastic discovery from phenotypic assays.

Multi-Endpoint Acute Toxicity Assessment of Organic Compounds Using Large-Scale Machine Learning Modeling

In recent years, alternative animal testing methods such as computational and machine learning approaches have become increasingly crucial for toxicity testing. However, the complexity and scarcity of available biomedical data challenge the development of predictive models. Combining nonlinear machine learning together with multicondition descriptors offers a solution for using data from various assays to create a robust model. This work applies multicondition descriptors (MCDs) to develop a QSTR (Quantitative Structure-Toxicity Relationship) model based on a large toxicity data set comprising more than 80,000 compounds and 59 different end points (122,572 data points). The prediction capabilities of developed single-task multi-end point machine learning models as well as a novel data analysis approach with the use of Convolutional Neural Networks (CNN) are discussed. The results show that using MCDs significantly improves the model and using them with CNN-1D yields the best result (R2train = 0.93, R2ext = 0.70). Several structural features showed a high level of contribution to the toxicity, including van der Waals surface area (VSA), number of nitrogen-containing fragments (nN+), presence of S-P fragments, ionization potential, and presence of C-N fragments. The developed models can be very useful tools to predict the toxicity of various compounds under different conditions, enabling quick toxicity assessment of new compounds.

Machine Learning Exploration of the Relationship Between Drugs and the Blood-Brain Barrier: Guiding Molecular Modification

OBJECTIVE

This study aimed to improve the efficiency of pharmacotherapy for CNS diseases by optimizing the ability of drug molecules to penetrate the Blood-Brain Barrier (BBB).

METHODS

We established qualitative and quantitative databases of the ADME properties of drugs and derived characteristic features of compounds with efficient BBB penetration. Using these insights, we developed four machine learning models to predict a drug's BBB permeability by assessing ADME properties and molecular topology. We then validated the models using the B3DB database. For acyclovir and ceftriaxone, we modified the Hydrogen Bond Donors and Acceptors, and evaluated the BBB permeability using the predictive model.

RESULTS

The machine learning models performed well in predicting BBB permeability on both internal and external validation sets. Reducing the number of Hydrogen Bond Donors and Acceptors generally improves BBB permeability. Modification only enhanced BBB penetration in the case of acyclovir and not ceftriaxone.

CONCLUSIONS

The machine learning models developed can accurately predict BBB permeability, and many drug molecules are likely to have increased BBB penetration if the number of Hydrogen Bond Donors and Acceptors are reduced. These findings suggest that molecular modifications can enhance the efficacy of CNS drugs and provide practical strategies for drug design and development. This is particularly relevant for improving drug penetration of the BBB.

Optimizing drug discovery using multitasking models for quantitative structure-biological effect relationships: an update of the literature

INTRODUCTION

Drug discovery has provided modern societies with the means to fight against many diseases. In this sense, computational methods have been at the forefront, playing an important role in rationalizing the search for novel drugs. Yet, tackling phenomena such as the multi-genic nature of diseases and drug resistance are limitations of the current computational methods. Multi-tasking models for quantitative structure-biological effect relationships (mtk-QSBER) have emerged to overcome such limitations.

AREAS COVERED

The present review describes an update on the fundamentals and applications of the mtk-QSBER models as tools to accelerate multiple stages/substages of the drug discovery process.

EXPERT OPINION

Computational approaches are extremely important for the rationalization of the search for novel and efficacious therapeutic agents. However, they need to focus more on the multi-target drug discovery paradigm. In this sense, mtk-QSBER models are particularly suited for multi-target drug discovery, offering encouraging opportunities across multiple therapeutic areas and scientific disciplines associated with drug discovery.

Machine learning in antibacterial discovery and development: A bibliometric and network analysis of research hotspots and trends

Machine learning (ML) methods are used in cheminformatics processes to predict the activity of an unknown drug and thus discover new potential antibacterial drugs. This article conducts a bibliometric study to analyse the contributions of leading authors, universities/organisations and countries in terms of productivity, citations and bibliographic linkage. A sample of 1596 Scopus documents for the period 2006-2022 is the basis of the study. In order to develop the analysis, bibliometrix R-Tool and VOSviewer software were used. We determined essential topics related to the application of ML in the field of antibacterial development (Computer model in antibacterial drug design, and Learning algorithms and systems for forecasting). We identified obsolete and saturated areas of research. At the same time, we proposed emerging topics according to the various analyses carried out on the corpus of published scientific literature (Title, abstract and keywords). Finally, the applied methodology contributed to building a broader and more specific "big picture" of ML research in antibacterial studies for the focus of future projects.