Merging Counter-Propagation and Back-Propagation Algorithms: Overcoming the Limitations of Counter-Propagation Neural Network Models.

Artificial neural networks (ANNs) are nowadays applied as the most efficient methods in the majority of machine learning approaches, including data-driven modeling for assessment of the toxicity of chemicals. We developed a combined neural network methodology that can be used in the scope of new approach methodologies (NAMs) assessing chemical or drug toxicity. Here, we present QSAR models for predicting the physical and biochemical properties of molecules of three different datasets: aqueous solubility, acute fish toxicity toward fat head minnow, and bio-concentration factors. A novel neural network modeling method is developed by combining two neural network algorithms, namely, the counter-propagation modeling strategy (CP-ANN) with the back-propagation-of-errors algorithm (BPE-ANN). The advantage is a short training time, robustness, and good interpretability through the initial CP-ANN part, while the extension with BPE-ANN improves the precision of predictions in the range between minimal and maximal property values of the training data, regardless of the number of neurons in both neural networks, either CP-ANN or BPE-ANN.

Cheminformatics and Machine Learning Approaches to Assess Aquatic Toxicity Profiles of Fullerene Derivatives.

Fullerene derivatives (FDs) are widely used in nanomaterials production, the pharmaceutical industry and biomedicine. In the present study, we focused on the potential toxic effects of FDs on the aquatic environment. First, we analyzed the binding affinity of 169 FDs to 10 human proteins (1D6U, 1E3K, 1GOS, 1GS4, 1H82, 1OG5, 1UOM, 2F9Q, 2J0D, 3ERT) obtained from the Protein Data Bank (PDB) and showing high similarity to proteins from aquatic species. Then, the binding activity of 169 FDs to the enzyme acetylcholinesterase (AChE)-as a known target of toxins in fathead minnows and Daphnia magna, causing the inhibition of AChE-was analyzed. Finally, the structural aquatic toxicity alerts obtained from ToxAlert were used to confirm the possible mechanism of action. Machine learning and cheminformatics tools were used to analyze the data. Counter-propagation artificial neural network (CPANN) models were used to determine key binding properties of FDs to proteins associated with aquatic toxicity. Predicting the binding affinity of unknown FDs using quantitative structure-activity relationship (QSAR) models eliminates the need for complex and time-consuming calculations. The results of the study show which structural features of FDs have the greatest impact on aquatic organisms and help prioritize FDs and make manufacturing decisions.

Recent Advances on P-Glycoprotein (ABCB1) Transporter Modelling with In Silico Methods.

ABC transporters play a critical role in both drug bioavailability and toxicity, and with the discovery of the P-glycoprotein (P-gp), this became even more evident, as it plays an important role in preventing intracellular accumulation of toxic compounds. Over the past 30 years, intensive studies have been conducted to find new therapeutic molecules to reverse the phenomenon of multidrug resistance (MDR) ), that research has found is often associated with overexpression of P-gp, the most extensively studied drug efflux transporter; in MDR, therapeutic drugs are prevented from reaching their targets due to active efflux from the cell. The development of P-gp inhibitors is recognized as a good way to reverse this type of MDR, which has been the subject of extensive studies over the past few decades. Despite the progress made, no effective P-gp inhibitors to reverse multidrug resistance are yet on the market, mainly because of their toxic effects. Computational studies can accelerate this process, and in silico models such as QSAR models that predict the activity of compounds associated with P-gp (or analogous transporters) are of great value in the early stages of drug development, along with molecular modelling methods, which provide a way to explain how these molecules interact with the ABC transporter. This review highlights recent advances in computational P-gp research, spanning the last five years to 2022. Particular attention is given to the use of machine-learning approaches, drug-transporter interactions, and recent discoveries of potential P-gp inhibitors that could act as modulators of multidrug resistance.

Structure-Function Relationships in the Human P-Glycoprotein (ABCB1): Insights from Molecular Dynamics Simulations.

P-Glycoprotein (P-gp) is a transmembrane protein belonging to the ATP binding cassette superfamily of transporters, and it is a xenobiotic efflux pump that limits intracellular drug accumulation by pumping compounds out of cells. P-gp contributes to a reduction in toxicity, and has broad substrate specificity. It is involved in the failure of many cancer and antiviral chemotherapies due to the phenomenon of multidrug resistance (MDR), in which the membrane transporter removes chemotherapeutic drugs from target cells. Understanding the details of the ligand-P-gp interaction is therefore critical for the development of drugs that can overcome the MDR phenomenon, for the early identification of P-gp substrates that will help us to obtain a more effective prediction of toxicity, and for the subsequent outdesign of substrate properties if needed. In this work, a series of molecular dynamics (MD) simulations of human P-gp (hP-gp) in an explicit membrane-and-water environment were performed to investigate the effects of binding different compounds on the conformational dynamics of P-gp. The results revealed significant differences in the behaviour of P-gp in the presence of active and non-active compounds within the binding pocket, as different patterns of movement were identified that could be correlated with conformational changes leading to the activation of the translocation mechanism. The predicted ligand-P-gp interactions are in good agreement with the available experimental data, as well as the estimation of the binding-free energies of the studied complexes, demonstrating the validity of the results derived from the MD simulations.

Homology Modeling of the Human P-glycoprotein (ABCB1) and Insights into Ligand Binding through Molecular Docking Studies.

The ABCB1 transporter also known as P-glycoprotein (P-gp) is a transmembrane protein belonging to the ATP binding cassette super-family of transporters; it is a xenobiotic efflux pump that limits intracellular drug accumulation by pumping the compounds out of cells. P-gp contributes to a decrease of toxicity and possesses broad substrate specificity. It is involved in the failure of numerous anticancer and antiviral chemotherapies due to the multidrug resistance (MDR) phenomenon, where it removes the chemotherapeutics out of the targeted cells. Understanding the details of the ligand-P-gp interaction is therefore crucial for the development of drugs that might overcome the MRD phenomenon and for obtaining a more effective prediction of the toxicity of certain compounds. In this work, an in silico modeling was performed using homology modeling and molecular docking methods with the aim of better understanding the ligand-P-gp interactions. Based on different mouse P-gp structural templates from the PDB repository, a 3D model of the human P-gp (hP-gp) was constructed by means of protein homology modeling. The homology model was then used to perform molecular docking calculations on a set of thirteen compounds, including some well-known compounds that interact with P-gp as substrates, inhibitors, or both. The sum of ranking differences (SRD) was employed for the comparison of the different scoring functions used in the docking calculations. A consensus-ranking scheme was employed for the selection of the top-ranked pose for each docked ligand. The docking results showed that a high number of π interactions, mainly π-sigma, π-alkyl, and π-π type of interactions, together with the simultaneous presence of hydrogen bond interactions contribute to the stability of the ligand-protein complex in the binding site. It was also observed that some interacting residues in hP-gp are the same when compared to those observed in a co-crystallized ligand (PBDE-100) with mouse P-gp (PDB ID: 4XWK). Our in silico approach is consistent with available experimental results regarding P-gp efflux transport assay; therefore it could be useful in the prediction of the role of new compounds in systemic toxicity.

Revisiting fish toxicity of active pharmaceutical ingredients: Mechanistic insights from integrated ligand-/structure-based assessments on acetylcholinesterase.

The release of active pharmaceutical ingredients (APIs) into the environment is of great concern for aquatic ecosystem as many of these chemicals are designed to exert biological activity. Hence, their impact on non-target organisms like fish would not be surprising. In this respect, we revisited fish toxicity data of pharmaceuticals to generate linear and non-linear quantitative structure-toxicity relationships (QSTRs). We predicted fish lethality data from the validated QSTR models for 120 APIs with no experimental fish toxicity data. Toxicity of APIs on aquatic organisms is not fully characterized. Therefore, to provide a mechanistic insight for the assessment of API's toxicity to fish, the outcome of the derived QSTR models was integrated with structure-based toxicophore and molecular docking studies, utilizing the biomarker enzyme acetylcholinesterase originating from fish Torpedo californica (TcAChE). Toxicophore virtual screening of 60 chemicals with pT > 0 identified 23 hits as potential TcAChE binders with binding free energies ranging from -6.5 to -12.9 kcal/mol. The TcAChE-ligand interaction analysis revealed a good nesting of all 23 hits within TcAChE binding site through establishing strong lipophilic and hydrogen bonding interactions with the surrounding key amino acid residues. Among the chemicals passing the criteria of our integrated approach, majority of APIs belong noticeably to the Central Nervous System class. The screened chemicals displayed not only comprehensive toxicophore coverage, but also strong binding affinities according to the docking calculations, mainly due to interactions with TcAChE's key amino acid residues Tyr121, Tyr130, Tyr334, Trp84, Phe290, Phe330, Phe331, Ser122, and Ser200. Moreover, we propose here that binding of pharmaceuticals to AChE might have a potential in triggering molecular initiating events for adverse outcome pathways (AOPs), which in turn can play an important role for future screening of APIs lacking fish lethality data.

Structural Analysis and Dynamic Processes of the Transmembrane Segment Inside Different Micellar Environments-Implications for the TM4 Fragment of the Bilitranslocase Protein.

The transmembrane (TM) proteins are gateways for molecular transport across the cell membrane that are often selected as potential targets for drug design. The bilitranslocase (BTL) protein facilitates the uptake of various anions, such as bilirubin, from the blood into the liver cells. As previously established, there are four hydrophobic transmembrane segments (TM1-TM4), which constitute the structure of the transmembrane channel of the BTL protein. In our previous studies, the 3D high-resolution structure of the TM2 and TM3 transmembrane fragments of the BTL in sodium dodecyl sulfate (SDS) micellar media were solved using Nuclear Magnetic Resonance (NMR) spectroscopy and molecular dynamics simulations (MD). The high-resolution 3D structure of the fourth transmembrane region (TM4) of the BTL was evaluated using NMR spectroscopy in two different micellar media, anionic SDS and zwitterionic DPC (dodecylphosphocholine). The presented experimental data revealed the existence of an α -helical conformation in the central part of the TM4 in both micellar media. In the case of SDS surfactant, the α -helical conformation is observed for the Pro258-Asn269 region. The use of the zwitterionic DPC micelle leads to the formation of an amphipathic α -helix, which is characterized by the extension of the central α -helix in the TM4 fragment to Phe257-Thr271. The complex character of the dynamic processes in the TM4 peptide within both surfactants was analyzed based on the relaxation data acquired on 15 N and 31 P isotopes. Contrary to previously published and present observations in the SDS micelle, the zwitterionic DPC environment leads to intensive low-frequency molecular dynamic processes in the TM4 fragment.

An In Silico Approach for Assessment of the Membrane Transporter Activities of Phenols: A Case Study Based on Computational Models of Transport Activity for the Transporter Bilitranslocase.

Phenols are the most abundant naturally accessible antioxidants present in a human normal diet. Since numerous beneficial applications of phenols as preventive agents in various diseases were revealed, the evaluation of phenols bioavailability is of high interest of researchers, consumers and drug manufacturers. The hydrophilic nature of phenols makes a cell membrane penetration difficult, which imply an alternative way of uptake via membrane transporters. However, the structural and functional data of membrane transporters are limited, thus the in silico modelling is really challenging and urgent tool in elucidation of transporter ligands. Focus of this research was a particular transporter bilitranslocase (BTL). BTL has a broad tissue expression (vascular endothelium, absorptive and excretory epithelia) and can transport wide variety of poly-aromatic compounds. With available BTL data (pKi [mmol/L] for 120 organic compounds) a robust and reliable QSAR models for BTL transport activity were developed and extrapolated on 300 phenolic compounds. For all compounds the transporter profiles were assessed and results show that dietary phenols and some drug candidates are likely to interact with BTL. Moreover, synopsis of predictions from BTL models and hits/predictions of 20 transporters from Metrabase and Chembench platforms were revealed. With such joint transporter analyses a new insights for elucidation of BTL functional role were acquired. Regarding limitation of models for virtual profiling of transporter interactions the computational approach reported in this study could be applied for further development of reliable in silico models for any transporter, if in vitro experimental data are available.

Multiclass Classifier for P-Glycoprotein Substrates, Inhibitors, and Non-Active Compounds.

P-glycoprotein (P-gp) is a transmembrane protein that actively transports a wide variety of chemically diverse compounds out of the cell. It is highly associated with the ADMET (absorption, distribution, metabolism, excretion and toxicity) properties of drugs/drug candidates and contributes to decreasing toxicity by eliminating compounds from cells, thereby preventing intracellular accumulation. Therefore, in the drug discovery and toxicological assessment process it is advisable to pay attention to whether a compound under development could be transported by P-gp or not. In this study, an in silico multiclass classification model capable of predicting the probability of a compound to interact with P-gp was developed using a counter-propagation artificial neural network (CP ANN) based on a set of 2D molecular descriptors, as well as an extensive dataset of 2512 compounds (1178 P-gp inhibitors, 477 P-gp substrates and 857 P-gp non-active compounds). The model provided a good classification performance, producing non error rate (NER) values of 0.93 for the training set and 0.85 for the test set, while the average precision (AvPr) was 0.93 for the training set and 0.87 for the test set. An external validation set of 385 compounds was used to challenge the model's performance. On the external validation set the NER and AvPr values were 0.70 for both indices. We believe that this in silico classifier could be effectively used as a reliable virtual screening tool for identifying potential P-gp ligands.

Discovery of (phenylureido)piperidinyl benzamides as prospective inhibitors of bacterial autolysin E from Staphylococcus aureus.

Autolysin E (AtlE) is a cell wall degrading enzyme that catalyzes the hydrolysis of the ß-1,4-glycosidic bond between the N-acetylglucosamine and N-acetylmuramic acid units of the bacterial peptidoglycan. Using our recently determined crystal structure of AtlE from Staphylococcus aureus and a combination of pharmacophore modeling, similarity search, and molecular docking, a series of (Phenylureido)piperidinyl benzamides were identified as potential binders and surface plasmon resonance (SPR) and saturation-transfer difference (STD) NMR experiments revealed that discovered compounds bind to AtlE in a lower micromolar range. (phenylureido)piperidinyl benzamides are the first reported non-substrate-like compounds that interact with this enzyme and enable further study of the interaction of small molecules with bacterial AtlE as potential inhibitors of this target.

Structural elucidation of transmembrane transporter protein bilitranslocase: conformational analysis of the second transmembrane region TM2 by molecular dynamics and NMR spectroscopy.

Membrane proteins represent about a third of the gene products in most organisms, as revealed by the genome sequencing projects. They account for up to two thirds of known drugable targets, which emphasizes their critical pharmaceutical importance. Here we present a study on bilitranslocase (BTL) (TCDB 2.A.65), a membrane protein primarily involved in the transport of bilirubin from blood to liver cells. Bilitranslocase has also been identified as a potential membrane transporter for cellular uptake of several drugs and due to its implication in drug uptake, it is extremely important to advance the knowledge about its 3D structure. However, at present, only a limited knowledge is available beyond the primary structure of BTL. It has been recently confirmed experimentally that one of the four computationally predicted transmembrane segments of bilitranslocase, TM3, has a helical structure with hydrophilic amino acid residues oriented towards one side, which is typical for transmembrane domains of membrane proteins. In this study we confirmed by the use of multidimensional NMR spectroscopy that the second transmembrane segment, TM2, also appears in a form of α-helix. The stability of this polypeptide chain was verified by molecular dynamics (MD) simulation in dipalmitoyl phosphatidyl choline (DPPC) and in sodium dodecyl sulfate (SDS) micelles. The two α-helices, TM2 corroborated in this study, and TM3 confirmed in our previous investigation, provide reasonable building blocks of a potential transmembrane channel for transport of bilirubin and small hydrophilic molecules, including pharmaceutically active compounds.

Prediction of antiprion activity of therapeutic agents with structure-activity models.

We have developed computational structure-activity models for the prediction of antiprion activity of compounds with known molecular structure. The aim is to apply the developed classification and predictive models in further drug design of antiprion therapeutics. The neural network models developed on the counter-propagation reinforcement learning strategy performed better than the linear regression models. The initial data set was composed of 461 compounds representing diverse groups of chemicals (derivatives of acridine, quinolone, Congo red, 2-aminopyridine-3,5-dicarbonitrile, styrylbenzoazole, 2,5-diamino-benzoquinone), which have been tested in comparable cell-screening assay studies for their activity against prion accumulation. Initially, we have designed a classification model for preliminary sorting of compounds into highly active, active, and inactive groups. Further, only the active compounds with IC50 less or equal to 10 µM were considered as the initial source of data. Altogether, 158 compounds were used to train the artificial neural network model for the estimation of the antiprion activity. The predictive ability of the model was significantly improved after selection of influential variables with genetic algorithm. The root- mean-squared error of the predicted pIC50 values for the external validation set (RMS EV) was slightly above 0.50 log units. A linear regression model, developed for the reasons of comparison, performed with a lower predictive ability (RMS EV 0.92 log units). The applicability domain of the models was assessed by a leverage and distance approach. The set of selected influential structural variables was further studied with the aim to get a better insight into the structural features of compounds potentially involved in disturbing of the prion-prion interactions.

Exploring the dynamics of the ABCB1 membrane transporter P-glycoprotein in the presence of ATP and active/non-active compounds through molecular dynamics simulations.

P-glycoprotein (Pgp) is a member of the ATP-binding cassette family of transporters that confers multidrug resistance to cancer cells and is actively involved in the pharmacokinetics and toxicokinetics of a big variety of drugs. Extensive studies have provided insights into the binding of many compounds, but the precise mechanism of translocation across the membrane remains unknown; in this context, the major challenge has been to understand the basis for its polyspecificity. In this study, molecular dynamics (MD) simulations of human P-gp (hP-gp) in an explicit membrane-and-water environment were performed to investigate the dynamic behavior of the transporter in the presence of different compounds (active and inactive) in the binding pocket and ATP molecules within the nucleotide binding domains (NBDs). The complexes studied involve four compounds: cyclosporin A (CSA), amiodarone (AMI), pamidronate (APD), and valproic acid (VPA). While CSA and AMI are known to interact with P-gp, APD and VPA do not. The results highlighted how the presence of ATP notably contributed to increased flexibility of key residues in NBD1 of active systems, indicating potential conformational changes activating the translocation mechanism. MD simulations reveal how these domains adapt and respond to the presence of different substrates, as well as the influence of ATP binding on their flexibility. Furthermore, distinctive behavior was observed in the presence of active and inactive compounds, particularly in the arrangement of ATP between NBDs, supporting the proposed nucleotide sandwich dimer mechanism for ATP binding. This study provides comprehensive insights into P-gp behavior with various ligands and ATP, offering implications for drug development, toxicity assessment and demonstrating the validity of the results derived from the MD simulations.

Common vertex matrix: a novel characterization of molecular graphs by counting.

We present a novel matrix representation of graphs based on the count of equal-distance common vertices to each pair of vertices in a graph. The element (i, j) of this matrix is defined as the number of vertices at the same distance from vertices (i, j). As illustrated on smaller alkanes, these novel matrices are very sensitive to molecular branching and the distribution of vertices in a graph. In particular, we show that ordered row sums of these novel matrices can facilitate solving graph isomorphism for acyclic graphs. This has been illustrated on all undecane isomers C11H24 having the same path counts (total of 25 molecules), on pair of graphs on 18 vertices having the same distance degree sequences (Slater's graphs), as well as two graphs on 21 vertices having identical several topological indices derived from information on distances between vertices.

On the centrality of vertices of molecular graphs.

For acyclic systems the center of a graph has been known to be either a single vertex of two adjacent vertices, that is, an edge. It has not been quite clear how to extend the concept of graph center to polycyclic systems. Several approaches to the graph center of molecular graphs of polycyclic graphs have been proposed in the literature. In most cases alternative approaches, however, while being apparently equally plausible, gave the same results for many molecules, but occasionally they differ in their characterization of molecular center. In order to reduce the number of vertices that would qualify as forming the center of the graph, a hierarchy of rules have been considered in the search for graph centers. We reconsidered the problem of "the center of a graph" by using a novel concept of graph theory, the vertex "weights," defined by counting the number of pairs of vertices at the same distance from the vertex considered. This approach gives often the same results for graph centers of acyclic graphs as the standard definition of graph center based on vertex eccentricities. However, in some cases when two nonequivalent vertices have been found as graph center, the novel approach can discriminate between the two. The same approach applies to cyclic graphs without additional rules to locate the vertex or vertices forming the center of polycyclic graphs, vertices referred to as central vertices of a graph. In addition, the novel vertex "weights," in the case of acyclic, cyclic, and polycyclic graphs can be interpreted as vertex centralities, a measure for how close or distant vertices are from the center or central vertices of the graph. Besides illustrating the centralities of a number of smaller polycyclic graphs, we also report on several acyclic graphs showing the same centrality values of their vertices.

Cluster-based molecular docking study for in silico identification of novel 6-fluoroquinolones as potential inhibitors against Mycobacterium tuberculosis.

A classical protein sequence alignment and homology modeling strategy were used for building three Mycobacterium tuberculosis-DNA gyrase protein models using the available topoII-DNA-6FQ crystal structure complexes originating from different organisms. The recently determined M. tuberculosis-DNA gyrase apoprotein structures and topoII-DNA-6FQ complexes were used for defining the 6-fluoroquinolones (6-FQs) binding pockets. The quality of the generated models was initially validated by docking of the cocrystallized ligands into their binding site, and subsequently by quantitative evaluation of their discriminatory performances (identification of active/inactive 6-FQs) for a set of 145 6-FQs with known biological activity values. The M. tuberculosis-DNA gyrase model with the highest estimated discriminatory power was selected and used afterwards in an additional molecular docking experiment on a mixed combinatorial set of 427 drug-like 6-FQ analogs for which the biological activity values were predicted using a prebuilt counter-propagation artificial neural network model. A novel three-level Boolean-based [T/F (true/false)] clustering algorithm was used to assess the generated binding poses: Level 1 (geometry properties assessment), Level 2 (score-based clustering and selection of the (T)-signed highly scored Level 1 poses), and Level 3 (activity-based clustering and selection of the most "active" (T)-signed Level 2 hits). The frequency analysis of occurrence of the fragments attached at R(1) and R(7) position of the (T)-signed 6-FQs selected in Level 3 revealed several novel attractive fragments and confirmed some previous findings. We believe that this methodology could be successfully used in establishing novel possible structure-activity relationship recommendations in the 6-FQs optimization, which could be of great importance in the current antimycobacterial hit-to-lead processes.

Mathematical characterization of protein transmembrane regions.

Graphical bioinformatics has paved a unique way of mathematical characterization of proteins and proteomic maps. The graphics representations and the corresponding mathematical descriptors have proved to be useful and have provided unique solutions to problems related to identification, comparisons, and analyses of protein sequences and proteomics maps. Based on sequence information alone, these descriptors are independent from physiochemical properties of amino acids and evolutionary information. In this work, we have presented invariants from amino acid adjacency matrix and decagonal isometries matrix as potential descriptors of protein sequences. Encoding protein sequences into amino acid adjacency matrix is already well established. We have shown its application in classification of transmembrane and nontransmembrane regions of membrane protein sequences. We have introduced the dodecagonal isometries matrix, which is a novel method of encoding protein sequences based on decagonal isometries group.

Quantitative Structure Activity/Toxicity Relationship through Neural Networks for Drug Discovery or Regulatory Use.

Quantitative structure - activity relationship (QSAR) modelling is widely used in medicinal chemistry and regulatory decision making. The large amounts of data collected in recent years in materials and life sciences projects provide a solid foundation for data-driven modelling approaches that have fostered the development of machine learning and artificial intelligence tools. An overview and discussion of the principles of QSAR modelling focus on the assembly and curation of data, computation of molecular descriptor, optimization, validation, and definition of the scope of the developed QSAR models. In this review, some examples of (QSAR) models based on artificial neural networks are given to demonstrate the effectiveness of nonlinear methods for extracting information from large data sets to classify new chemicals and predict their biological properties.

Exploration of structural requirements for azole chemicals towards human aromatase CYP19A1 activity: Classification modeling, structure-activity relationships and read-across study.

Human aromatase, also called CYP19A1, plays a major role in the conversion of androgens into estrogens. Inhibition of aromatase is an important target for estrogen receptor (ER)-responsive breast cancer therapy. Use of azole compounds as aromatase inhibitors is widespread despite their low selectivity. A toxicological evaluation of commonly used azole-based drugs and agrochemicals with respect to CYP19A1 is currently requested by the European Union- Registration, Evaluation, Authorization and Restriction of Chemicals (EU-REACH) regulations due to their potential as endocrine disruptors. In this connection, identification of structural alerts (SAs) is an effective strategy for the toxicological assessment and safe drug design. The present study describes the identification of SAs of azole-based chemicals as guiding experts to predict the aromatase activity. Total 21 SAs associated with aromatase activity were extracted from dataset of 326 azole-based drugs/chemicals obtained from Tox21 library. A cross-validated classification model having high accuracy (error rate 5%) was proposed which can precisely classify azole chemicals into active/inactive toward aromatase. In addition, mechanistic details and toxicological properties (agonism/antagonism) of azoles with respect to aromatase were explored by comparing active and inactive chemicals using structure-activity relationships (SAR). Lastly, few structural alerts were applied to form chemical categories for read-across applications.

Probabilistic modelling of developmental neurotoxicity based on a simplified adverse outcome pathway network.

In a century where toxicology and chemical risk assessment are embracing alternative methods to animal testing, there is an opportunity to understand the causal factors of neurodevelopmental disorders such as learning and memory disabilities in children, as a foundation to predict adverse effects. New testing paradigms, along with the advances in probabilistic modelling, can help with the formulation of mechanistically-driven hypotheses on how exposure to environmental chemicals could potentially lead to developmental neurotoxicity (DNT). This investigation aimed to develop a Bayesian hierarchical model of a simplified AOP network for DNT. The model predicted the probability that a compound induces each of three selected common key events (CKEs) of the simplified AOP network and the adverse outcome (AO) of DNT, taking into account correlations and causal relations informed by the key event relationships (KERs). A dataset of 88 compounds representing pharmaceuticals, industrial chemicals and pesticides was compiled including physicochemical properties as well as in silico and in vitro information. The Bayesian model was able to predict DNT potential with an accuracy of 76%, classifying the compounds into low, medium or high probability classes. The modelling workflow achieved three further goals: it dealt with missing values; accommodated unbalanced and correlated data; and followed the structure of a directed acyclic graph (DAG) to simulate the simplified AOP network. Overall, the model demonstrated the utility of Bayesian hierarchical modelling for the development of quantitative AOP (qAOP) models and for informing the use of new approach methodologies (NAMs) in chemical risk assessment.