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.

Design and Evaluation of Nanoscale Materials with Programmed Responsivity towards Epigenetic Enzymes.

Self-assembled materials capable of modulating their assembly properties in response to specific enzymes play a pivotal role in advancing 'intelligent' encapsulation platforms for biotechnological applications. Here, we introduce a previously unreported class of synthetic nanomaterials that programmatically interact with histone deacetylase (HDAC) as the triggering stimulus for disassembly. These nanomaterials consist of co-polypeptides comprising poly (acetyl L-lysine) and poly(ethylene glycol) blocks. Under neutral pH conditions, they self-assemble into particles. However, their stability is compromised upon exposure to HDACs, depending on enzyme concentration and exposure time. Our investigation, utilizing HDAC8 as the model enzyme, revealed that the primary mechanism behind disassembly involves a decrease in amphiphilicity within the block copolymer due to the deacetylation of lysine residues within the particles' hydrophobic domains. To elucidate the response mechanism, we encapsulated a fluorescent dye within these nanoparticles. Upon incubation with HDAC, the nanoparticle structure collapsed, leading to controlled release of the dye over time. Notably, this release was not triggered by denatured HDAC8, other proteolytic enzymes like trypsin, or the co-presence of HDAC8 and its inhibitor. We further demonstrated the biocompatibility and cellular effects of these materials and conducted a comprehensive computational study to unveil the possible interaction mechanism between enzymes and particles. By drawing parallels to the mechanism of naturally occurring histone proteins, this research represents a pioneering step toward developing functional materials capable of harnessing the activity of epigenetic enzymes such as HDACs.

Combined Machine Learning, Computational, and Experimental Analysis of the Iridium(III) Complexes with Red to Near-Infrared Emission.

Various coordination complexes have been the subject of experimental and theoretical studies in recent decades because of their fascinating photophysical properties. In this work, a combined experimental and computational approach was applied to investigate the optical properties of monocationic Ir(III) complexes. An interpretative machine learning-based quantitative structure-property relationship (ML/QSPR) model was successfully developed that could reliably predict the emission wavelength of the Ir(III) complexes and provide a foundation for the theoretical evaluation of the optical properties of Ir(III) complexes. A hypothesis was proposed to explain the differences in the emission wavelengths between structurally different individual Ir(III) complexes. The efficacy of the developed model was demonstrated by high R2 values of 0.84 and 0.87 for the training and test sets, respectively. It is worth noting that a relationship between the N-N distance in the diimine ligands of the Ir(III) complexes and emission wavelengths is detected. This effect is most probably associated with a degree of distortion in the octahedral geometry of the complexes, resulting in a perturbed ligand field. This combined experimental and computational approach shows great potential for the rational design of new Ir(III) complexes with the desired optical properties. Moreover, the developed methodology could be extended to other transition-metal complexes.

Stability Constant and Potentiometric Sensitivity of Heavy Metal-Organic Fluorescent Compound Complexes: QSPR Models for Prediction and Design of Novel Coumarin-like Ligands.

Industrial wastewater often consists of toxic chemicals and pollutants, which are extremely harmful to the environment. Heavy metals are toxic chemicals and considered one of the major hazards to the aquatic ecosystem. Analytical techniques, such as potentiometric methods, are some of the methods to detect heavy metals in wastewaters. In this work, the quantitative structure-property relationship (QSPR) was applied using a range of machine learning techniques to predict the stability constant (logßML) and potentiometric sensitivity (PSML) of 200 ligands in complexes with the heavy metal ions Cu2+, Cd2+, and Pb2+. In result, the logßML models developed for four ions showed good performance with square correlation coefficients (R2) ranging from 0.80 to 1.00 for the training and 0.72 to 0.85 for the test sets. Likewise, the PSML displayed acceptable performance with an R2 of 0.87 to 1.00 for the training and 0.73 to 0.95 for the test sets. By screening a virtual database of coumarin-like structures, several new ligands bearing the coumarin moiety were identified. Three of them, namely NEW02, NEW03, and NEW07, showed very good sensitivity and stability in the metal complexes. Subsequent quantum-chemical calculations, as well as physicochemical/toxicological profiling were performed to investigate their metal-binding ability and developability of the designed sensors. Finally, synthesis schemes are proposed to obtain these three ligands with major efficiency from simple resources. The three coumarins designed clearly demonstrated capability to be suitable as good florescent chemosensors towards heavy metals. Overall, the computational methods applied in this study showed a very good performance as useful tools for designing novel fluorescent probes and assessing their sensing abilities.

Development of Activity Rules and Chemical Fragment Design for In Silico Discovery of AChE and BACE1 Dual Inhibitors against Alzheimer's Disease.

Multi-target drug development has become an attractive strategy in the discovery of drugs to treat of Alzheimer's disease (AzD). In this study, for the first time, a rule-based machine learning (ML) approach with classification trees (CT) was applied for the rational design of novel dual-target acetylcholinesterase (AChE) and ß-site amyloid-protein precursor cleaving enzyme 1 (BACE1) inhibitors. Updated data from 3524 compounds with AChE and BACE1 measurements were curated from the ChEMBL database. The best global accuracies of training/external validation for AChE and BACE1 were 0.85/0.80 and 0.83/0.81, respectively. The rules were then applied to screen dual inhibitors from the original databases. Based on the best rules obtained from each classification tree, a set of potential AChE and BACE1 inhibitors were identified, and active fragments were extracted using Murcko-type decomposition analysis. More than 250 novel inhibitors were designed in silico based on active fragments and predicted AChE and BACE1 inhibitory activity using consensus QSAR models and docking validations. The rule-based and ML approach applied in this study may be useful for the in silico design and screening of new AChE and BACE1 dual inhibitors against AzD.

Exploring proteasome inhibition using atomic weighted vector indices and machine learning approaches.

Ubiquitin-proteasome system (UPS) is a highly regulated mechanism of intracellular protein degradation and turnover. The UPS is involved in different biological activities, such as the regulation of gene transcription and cell cycle. Several researchers have applied cheminformatics and artificial intelligence methods to study the inhibition of proteasomes, including the prediction of UPP inhibitors. Following this idea, we applied a new tool for obtaining molecular descriptors (MDs) for modeling proteasome Inhibition in terms of EC50 (µmol/L), in which a set of new MDs called atomic weighted vectors (AWV) and several prediction algorithms were used in cheminformatics studies. In the manuscript, a set of descriptors based on AWV are presented as datasets for training different machine learning techniques, such as linear regression, multiple linear regression (MLR), random forest (RF), K-nearest neighbors (IBK), multi-layer perceptron, best-first search, and genetic algorithm. The results suggest that these atomic descriptors allow adequate modeling of proteasome inhibitors despite artificial intelligence techniques, as a variant to build efficient models for the prediction of inhibitory activity.

In Silico Prediction of the Toxicity of Nitroaromatic Compounds: Application of Ensemble Learning QSAR Approach.

In this work, a dataset of more than 200 nitroaromatic compounds is used to develop Quantitative Structure-Activity Relationship (QSAR) models for the estimation of in vivo toxicity based on 50% lethal dose to rats (LD50). An initial set of 4885 molecular descriptors was generated and applied to build Support Vector Regression (SVR) models. The best two SVR models, SVR_A and SVR_B, were selected to build an Ensemble Model by means of Multiple Linear Regression (MLR). The obtained Ensemble Model showed improved performance over the base SVR models in the training set (R2 = 0.88), validation set (R2 = 0.95), and true external test set (R2 = 0.92). The models were also internally validated by 5-fold cross-validation and Y-scrambling experiments, showing that the models have high levels of goodness-of-fit, robustness and predictivity. The contribution of descriptors to the toxicity in the models was assessed using the Accumulated Local Effect (ALE) technique. The proposed approach provides an important tool to assess toxicity of nitroaromatic compounds, based on the ensemble QSAR model and the structural relationship to toxicity by analyzed contribution of the involved descriptors.

The Hydrolysis Rate of Paraoxonase-1 Q and R Isoenzymes: An In Silico Study Based on In Vitro Data.

Human serum paraoxonase-1 (PON1) is an important hydrolase-type enzyme found in numerous tissues. Notably, it can exist in two isozyme-forms, Q and R, that exhibit different activities. This study presents an in silico (QSAR, Docking, MD and QM/MM) study of a set of compounds on the activity towards the PON1 isoenzymes (QPON1 and RPON1). Different rates of reaction for the Q and R isoenzymes were analyzed by modelling the effect of Q192R mutation on active sites. It was concluded that the Q192R mutation is not even close to the active site, while it is still changing the geometry of it. Using the combined genetic algorithm with multiple linear regression (GA-MLR) technique, several QSAR models were developed and relative activity rates of the isozymes of PON1 explained. From these, two QSAR models were selected, one each for the QPON1 and RPON1. Best selected models are four-variable MLR models for both Q and R isozymes with squared correlation coefficient R2 values of 0.87 and 0.83, respectively. In addition, the applicability domain of the models was analyzed based on the Williams plot. The results were discussed in the light of the main factors that influence the hydrolysis activity of the PON1 isozymes.

A Fuzzy System Classification Approach for QSAR Modeling of α- Amylase and α-Glucosidase Inhibitors.

INTRODUCTION: This report proposes the application of a new Machine Learning algorithm called Fuzzy Unordered Rules Induction Algorithm (FURIA)-C in the classification of druglike compounds with antidiabetic inhibitory ability toward the main two pharmacological targets: α-amylase and α-glucosidase. METHODS: The two obtained QSAR models were tested for classification capability, achieving satisfactory accuracy scores of 94.5% and 96.5%, respectively. Another important outcome was to achieve various α-amylase and α-glucosidase fuzzy rules with high Certainty Factor values. Fuzzyrules derived from the training series and active classification rules were interpreted. An important external validation step, comparing our method with those previously reported, was also included. RESULTS: The Holm's test comparison showed significant differences (p-value<0.05) between FURIA-C, Linear Discriminating Analysis (LDA), and Bayesian Networks, the former beating the two latter according to the relative ranking score of the Holm's test. CONCLUSION: From these results, the FURIA-C algorithm could be used as a cutting-edge technique to predict (classify or screen) the α-amylase and α-glucosidase inhibitory activity of new compounds and hence speed up the discovery of new potent multi-target antidiabetic agents.

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

Antibacterial drugs (AD) change the metabolic status of bacteria, contributing to bacterial death. However, antibiotic resistance and the emergence of multidrug-resistant bacteria increase interest in understanding metabolic network (MN) mutations and the interaction of AD vs MN. In this study, we employed the IFPTML = Information Fusion (IF) + Perturbation Theory (PT) + Machine Learning (ML) algorithm on a huge dataset from the ChEMBL database, which contains >155,000 AD assays vs >40 MNs of multiple bacteria species. We built a linear discriminant analysis (LDA) and 17 ML models centered on the linear index and based on atoms to predict antibacterial compounds. The IFPTML-LDA model presented the following results for the training subset: specificity (Sp) = 76% out of 70,000 cases, sensitivity (Sn) = 70%, and Accuracy (Acc) = 73%. The same model also presented the following results for the validation subsets: Sp = 76%, Sn = 70%, and Acc = 73.1%. Among the IFPTML nonlinear models, the k nearest neighbors (KNN) showed the best results with Sn = 99.2%, Sp = 95.5%, Acc = 97.4%, and Area Under Receiver Operating Characteristic (AUROC) = 0.998 in training sets. In the validation series, the Random Forest had the best results: Sn = 93.96% and Sp = 87.02% (AUROC = 0.945). The IFPTML linear and nonlinear models regarding the ADs vs MNs have good statistical parameters, and they could contribute toward finding new metabolic mutations in antibiotic resistance and reducing time/costs in antibacterial drug research.

Predicting Metabolic Reaction Networks with Perturbation-Theory Machine Learning (PTML) Models.

BACKGROUND: Checking the connectivity (structure) of complex Metabolic Reaction Networks (MRNs) models proposed for new microorganisms with promising properties is an important goal for chemical biology. OBJECTIVE: In principle, we can perform a hand-on checking (Manual Curation). However, this is a challenging task due to the high number of combinations of pairs of nodes (possible metabolic reactions). RESULTS: The CPTML linear model obtained using the LDA algorithm is able to discriminate nodes (metabolites) with the correct assignation of reactions from incorrect nodes with values of accuracy, specificity, and sensitivity in the range of 85-100% in both training and external validation data series. METHODS: In this work, we used Combinatorial Perturbation Theory and Machine Learning techniques to seek a CPTML model for MRNs >40 organisms compiled by Barabasis' group. First, we quantified the local structure of a very large set of nodes in each MRN using a new class of node index called Markov linear indices fk. Next, we calculated CPT operators for 150000 combinations of query and reference nodes of MRNs. Last, we used these CPT operators as inputs of different ML algorithms. CONCLUSION: Meanwhile, PTML models based on Bayesian network, J48-Decision Tree and Random Forest algorithms were identified as the three best non-linear models with accuracy greater than 97.5%. The present work opens the door to the study of MRNs of multiple organisms using PTML models.

Beyond model interpretability using LDA and decision trees for α-amylase and α-glucosidase inhibitor classification studies.

In this report are used two data sets involving the main antidiabetic enzyme targets α-amylase and α-glucosidase. The prediction of α-amylase and α-glucosidase inhibitory activity as antidiabetic is carried out using LDA and classification trees (CT). A large data set of 640 compounds for α-amylase and 1546 compounds in the case of α-glucosidase are selected to develop the tree model. In the case of CT-J48 have the better classification model performances for both targets with values above 80%-90% for the training and prediction sets, correspondingly. The best model shows an accuracy higher than 95% for training set; the model was also validated using 10-fold cross-validation procedure and through a test set achieving accuracy values of 85.32% and 86.80%, correspondingly. Additionally, the obtained model is compared with other approaches previously published in the international literature showing better results. Finally, we can say that the present results provided a double-target approach for increasing the estimation of antidiabetic chemicals identification aimed by double-way workflow in virtual screening pipelines.

In Silico Assessment of ADME Properties: Advances in Caco-2 Cell Monolayer Permeability Modeling.

One of the main goals of in silico Caco-2 cell permeability models is to identify those drug substances with high intestinal absorption in human (HIA). For more than a decade, several in silico Caco-2 models have been made, applying a wide range of modeling techniques; nevertheless, their capacity for intestinal absorption extrapolation is still doubtful. There are three main problems related to the modest capacity of obtained models, including the existence of inter- and/or intra-laboratory variability of recollected data, the influence of the metabolism mechanism, and the inconsistent in vitro-in vivo correlation (IVIVC) of Caco-2 cell permeability. This review paper intends to sum up the recent advances and limitations of current modeling approaches, and revealed some possible solutions to improve the applicability of in silico Caco-2 permeability models for absorption property profiling, taking into account the above-mentioned issues.

Learning from Multiple Classifier Systems: Perspectives for Improving Decision Making of QSAR Models in Medicinal Chemistry.

Quantitative Structure - Activity Relationship (QSAR) modeling has been widely used in medicinal chemistry and computational toxicology for many years. Today, as the amount of chemicals is increasing dramatically, QSAR methods have become pivotal for the purpose of handling the data, identifying a decision, and gathering useful information from data processing. The advances in this field have paved a way for numerous alternative approaches that require deep mathematics in order to enhance the learning capability of QSAR models. One of these directions is the use of Multiple Classifier Systems (MCSs) that potentially provide a means to exploit the advantages of manifold learning through decomposition frameworks, while improving generalization and predictive performance. In this paper, we presented MCS as a next generation of QSAR modeling techniques and discuss the chance to mining the vast number of models already published in the literature. We systematically revisited the theoretical frameworks of MCS as well as current advances in MCS application for QSAR practice. Furthermore, we illustrated our idea by describing ensemble approaches on modeling histone deacetylase (HDACs) inhibitors. We expect that our analysis would contribute to a better understanding about MCS application and its future perspectives for improving the decision making of QSAR models.

Carbon Nanotubes' Effect on Mitochondrial Oxygen Flux Dynamics: Polarography Experimental Study and Machine Learning Models using Star Graph Trace Invariants of Raman Spectra.

This study presents the impact of carbon nanotubes (CNTs) on mitochondrial oxygen mass flux (Jm) under three experimental conditions. New experimental results and a new methodology are reported for the first time and they are based on CNT Raman spectra star graph transform (spectral moments) and perturbation theory. The experimental measures of Jm showed that no tested CNT family can inhibit the oxygen consumption profiles of mitochondria. The best model for the prediction of Jm for other CNTs was provided by random forest using eight features, obtaining test R-squared (R²) of 0.863 and test root-mean-square error (RMSE) of 0.0461. The results demonstrate the capability of encoding CNT information into spectral moments of the Raman star graphs (SG) transform with a potential applicability as predictive tools in nanotechnology and material risk assessments.

A Simple Method to Predict Blood-Brain Barrier Permeability of Drug- Like Compounds Using Classification Trees.

BACKGROUND: To know the ability of a compound to penetrate the blood-brain barrier (BBB) is a challenging task; despite the numerous efforts realized to predict/measure BBB passage, they still have several drawbacks. METHOD: The prediction of the permeability through the BBB is carried out using classification trees. A large data set of 497 compounds (recently published) is selected to develop the tree model. RESULTS: The best model shows an accuracy higher than 87.6% for training set; the model was also validated using 10-fold cross-validation procedure and through a test set achieving accuracy values of 86.1% and 87.9%, correspondingly. We give a brief explanation, in structural terms, of how our model describes the passage of molecules through the BBB. Additionally, the obtained model is compared with other approaches previously published in the international literature showing better results. CONCLUSION: Finally, we can say that, the present model would be a valuable tool in the early stages of drug discovery process of neuropharmaceuticals.

Prediction of acute toxicity of phenol derivatives using multiple linear regression approach for Tetrahymena pyriformis contaminant identification in a median-size database.

In this article, the modeling of inhibitory grown activity against Tetrahymena pyriformis is described. The 0-2D Dragon descriptors based on structural aspects to gain some knowledge of factors influencing aquatic toxicity are mainly used. Besides, it is done by some enlarged data of phenol derivatives described for the first time and composed of 358 chemicals. It overcomes the previous datasets with about one hundred compounds. Moreover, the results of the model evaluation by the parameters in the training, prediction and validation give adequate results comparable with those of the previous works. The more influential descriptors included in the model are: X3A, MWC02, MWC10 and piPC03 with positive contributions to the dependent variable; and MWC09, piPC02 and TPC with negative contributions. In a next step, a median-size database of nearly 8000 phenolic compounds extracted from ChEMBL was evaluated with the quantitative-structure toxicity relationship (QSTR) model developed providing some clues (SARs) for identification of ecotoxicological compounds. The outcome of this report is very useful to screen chemical databases for finding the compounds responsible of aquatic contamination in the biomarker used in the current work.

Prediction of Aquatic Toxicity of Benzene Derivatives to Tetrahymena pyriformis According to OECD Principles.

BACKGROUND: Many QSAR studies have been developed to predict acute toxicity over several biomarkers like Pimephales promelas, Daphnia magna and Tetrahymena pyriformis. Regardless of the progress made in this field there are still some gaps to be resolved such as the prediction of aquatic toxicity over the protozoan T. pyriformis still lack a QSAR study focused in accomplish the OECD principles. METHODS: Atom-based quadratic indices are used to obtain quantitative structure-activity relationship (QSAR) models for the prediction of aquatic toxicity. Our models agree with the principles required by the OECD for QSAR models to regulatory purposes. The database employed consists of 392 substituted benzenes with toxicity values measured in T. pyriformis (defined endpoint), was divided using cluster analysis in two series (training and test sets). RESULTS: We obtain (with an unambiguous algorithm) two good multiple linear regression models for non-stochastic (R2=0.807 and s=0.334) and stochastic (R2=0.817 and s=0.321), quadratic indices. The models were internally validated using leave-one-out, bootstrapping as well as Y-scrambling experiments. We also perform an external validation using the test set, achieving values of R2 pred values of 0.754 and 0.760, showing that our models have appropriate measures of goodness- of-fit, robustness and predictivity. Moreover, we define a domain of applicability for our best models. CONCLUSION: The achieved results demonstrated that, the atom-based quadratic indices could provide an attractive alternative to the experiments currently used for determining toxicity, which are costly and time-consuming.

Multi-output Model with Box-Jenkins Operators of Quadratic Indices for Prediction of Malaria and Cancer Inhibitors Targeting Ubiquitin- Proteasome Pathway (UPP) Proteins.

The ubiquitin-proteasome pathway (UPP) is the primary degradation system of short-lived regulatory proteins. Cellular processes such as the cell cycle, signal transduction, gene expression, DNA repair and apoptosis are regulated by this UPP and dysfunctions in this system have important implications in the development of cancer, neurodegenerative, cardiac and other human pathologies. UPP seems also to be very important in the function of eukaryote cells of the human parasites like Plasmodium falciparum, the causal agent of the neglected disease Malaria. Hence, the UPP could be considered as an attractive target for the development of compounds with Anti-Malarial or Anti-cancer properties. Recent online databases like ChEMBL contains a larger quantity of information in terms of pharmacological assay protocols and compounds tested as UPP inhibitors under many different conditions. This large amount of data give new openings for the computer-aided identification of UPP inhibitors, but the intrinsic data diversity is an obstacle for the development of successful classifiers. To solve this problem here we used the Bob-Jenkins moving average operators and the atom-based quadratic molecular indices calculated with the software TOMOCOMD-CARDD (TC) to develop a quantitative model for the prediction of the multiple outputs in this complex dataset. Our multi-target model can predict results for drugs against 22 molecular or cellular targets of different organisms with accuracies above 70% in both training and validation sets.

Bond-based bilinear indices for computational discovery of novel trypanosomicidal drug-like compounds through virtual screening.

Two-dimensional bond-based bilinear indices and linear discriminant analysis are used in this report to perform a quantitative structure-activity relationship study to identify new trypanosomicidal compounds. A data set of 440 organic chemicals, 143 with antitrypanosomal activity and 297 having other clinical uses, is used to develop the theoretical models. Two discriminant models, computed using bond-based bilinear indices, are developed and both show accuracies higher than 86% for training and test sets. The stochastic model correctly indentifies nine out of eleven compounds of a set of organic chemicals obtained from our synthetic collaborators. The in vitro antitrypanosomal activity of this set against epimastigote forms of Trypanosoma cruzi is assayed. Both models show a good agreement between theoretical predictions and experimental results. Three compounds showed IC50 values for epimastigote elimination (AE) lower than 50 µM, while for the benznidazole the IC50 = 54.7 µM which was used as reference compound. The value of IC50 for cytotoxicity of these compounds is at least 5 times greater than their value of IC50 for AE. Finally, we can say that, the present algorithm constitutes a step forward in the search for efficient ways of discovering new antitrypanosomal compounds.