CORAL: QSPR model of water solubility based on local and global SMILES attributes

In silico Prediction of Skin Sensitization: Quo vadis?

Skin direct contact with chemical or physical substances is predisposed to allergic contact dermatitis (ACD), producing various allergic reactions, namely rash, blister, or itchy, in the contacted skin area. ACD can be triggered by various extremely complicated adverse outcome pathways (AOPs) remains to be causal for biosafety warrant. As such, commercial products such as ointments or cosmetics can fulfill the topically safe requirements in animal and non-animal models including allergy. Europe, nevertheless, has banned animal tests for the safety evaluations of cosmetic ingredients since 2013, followed by other countries. A variety of non-animal in vitro tests addressing different key events of the AOP, the direct peptide reactivity assay (DPRA), KeratinoSens™, LuSens and human cell line activation test h-CLAT and U-SENS™ have been developed and were adopted in OECD test guideline to identify the skin sensitizers. Other methods, such as the SENS-IS are not yet fully validated and regulatorily accepted. A broad spectrum of in silico models, alternatively, to predict skin sensitization have emerged based on various animal and non-animal data using assorted modeling schemes. In this article, we extensively summarize a number of skin sensitization predictive models that can be used in the biopharmaceutics and cosmeceuticals industries as well as their future perspectives, and the underlined challenges are also discussed.

Cytotoxicity of quantum dots: Use of quasiSMILES in development of reliable models with index of ideality of correlation and the consensus modelling

The assessment of cytotoxicity of quantum dots is very essential for environmental and health risk analysis. In the present work we have modelled HeLa cell cytotoxicity of sixty one CdSe quantum dots with ZnS shell as a function of its experimental conditions and molecular construction using quasiSMILES representations. The index of ideality of correlation helps in the building of ten statistically significant models having good fitting ability with value of R² ranging from 0.8414 to 0.9609 for the training set. The split 5 model is rated as the best model with values of R², Q²_F1, Q²_F2 and Q²_F3 as 0.8964, 0.8267, 0.8264 and 0.8777 respectively for the calibration set. The extraction of features causing increase and decrease of cytotoxicity of quantum dots indicates importance of neutral surface charge, surface modified with protein, 72 h exposure time, combination of MTT assay with surface protein in decreasing the cytotoxicity. Amphiphilic polymer, polyol ligand with neutral charge, 0.5 - 0.6 nm quantum dot diameter with lipid ligand and unmodified positively charged surface are grouped in toxicity enhancer features. Further, consensus modelling using split 5 and 8 patterns enhances the prediction quality by increasing the R²_val to 0.9361 and 0.9656 respectively.

The Monte Carlo Method as a Tool to Build up Predictive QSPR/QSAR

Background:

The Monte Carlo method has a wide application in various scientific researches. For the development of predictive models in a form of the quantitative structure-property / activity relationships (QSPRs/QSARs), the Monte Carlo approach also can be useful. The CORAL software provides the Monte Carlo calculations aimed to build up QSPR/QSAR models for different endpoints.

Methods:

Molecular descriptors are a mathematical function of so-called correlation weights of various molecular features. The numerical values of the correlation weights give the maximal value of a target function. The target function leads to a correlation between endpoint and optimal descriptor for the visible training set. The predictive potential of the model is estimated with the validation set, i.e. compounds that are not involved in the process of building up the model.

Results:

The approach gave quite good models for a large number of various physicochemical, biochemical, ecological, and medicinal endpoints. Bibliography and basic statistical characteristics of several CORAL models are collected in the present review. In addition, the extended version of the approach for more complex systems (nanomaterials and peptides), where behaviour of systems is defined by a group of conditions besides the molecular structure is demonstrated.

Conclusion:

The Monte Carlo technique available via the CORAL software can be a useful and convenient tool for the QSPR/QSAR analysis.

QSPR studies on water solubility, octanol-water partition coefficient and vapour pressure of pesticides

The assessment of the environmental fate and (eco)toxicological effects of pesticide compounds is of crucial importance. The present review is focused on Quantitative Structure-Property Relationships (QSPR) applications on three environmentally relevant physicochemical properties of pesticides, which can be used for assessing their environmental partition and transport, as well as exposure potential namely water solubility, octanol-water partition coefficient and vapour pressure. This article revises various interesting QSPR applications with special emphasis on studies developed during the 2009-2019 period.

Whether the Validation of the Predictive Potential of Toxicity Models is a Solved Task?

Different kinds of biological activities are defined by complex biochemical interactions, which are termed as a "mathematical function" not only of the molecular structure but also for some additional circumstances, such as physicochemical conditions, interactions via energy and information effects between a substance and organisms, organs, cells. These circumstances lead to the great complexity of prediction for biochemical endpoints, since all "details" of corresponding phenomena are practically unavailable for the accurate registration and analysis. Researchers have not a possibility to carry out and analyse all possible ways of the biochemical interactions, which define toxicological or therapeutically attractive effects via direct experiment. Consequently, a compromise, i.e. the development of predictive models of the above phenomena, becomes necessary. However, the estimation of the predictive potential of these models remains a task that is solved only partially. This mini-review presents a collection of attempts to be used for the above-mentioned task, two special statistical indices are proposed, which may be a measure of the predictive potential of models. These indices are (i) Index of Ideality of Correlation; and (ii) Correlation Contradiction Index.

Descriptor Free QSAR Modeling Using Deep Learning With Long Short-Term Memory Neural Networks

Current practice of building QSAR models usually involves computing a set of descriptors for the training set compounds, applying a descriptor selection algorithm and finally using a statistical fitting method to build the model. In this study, we explored the prospects of building good quality interpretable QSARs for big and diverse datasets, without using any pre-calculated descriptors. We have used different forms of Long Short-Term Memory (LSTM) neural networks to achieve this, trained directly using either traditional SMILES codes or a new linear molecular notation developed as part of this work. Three endpoints were modeled: Ames mutagenicity, inhibition of P. falciparum Dd2 and inhibition of Hepatitis C Virus, with training sets ranging from 7,866 to 31,919 compounds. To boost the interpretability of the prediction results, attention-based machine learning mechanism, jointly with a bidirectional LSTM was used to detect structural alerts for the mutagenicity data set. Traditional fragment descriptor-based models were used for comparison. As per the results of the external and cross-validation experiments, overall prediction accuracies of the LSTM models were close to the fragment-based models. However, LSTM models were superior in predicting test chemicals that are dissimilar to the training set compounds, a coveted quality of QSAR models in real world applications. In summary, it is possible to build QSAR models using LSTMs without using pre-computed traditional descriptors, and models are far from being "black box." We wish that this study will be helpful in bringing large, descriptor-less QSARs to mainstream use.

Conformation-independent quantitative structure-property relationships study on water solubility of pesticides

Water solubility is a key physicochemical parameter in pesticide control and regulation, although sometimes its experimental determination is not an easy task. In this study, we present Quantitative Structure-Property Relationships (QSPRs) for predicting the water solubility at 20 °C of 1211 approved heterogeneous pesticide compounds, collected from the online Pesticides Properties Data Base (PPDB). Validated and generally applicable Multivariable Linear Regression (MLR) models were established, including molecular descriptors carrying constitutional and topological aspects of the analyzed compounds. The most representative descriptors were selected from the exploration of a large number of about 18,000 structural variables. A hybrid approach that involves a molecular descriptor, a fingerprint, and a flexible descriptor showed the best predictive performance.

Quasi-QSAR for predicting the cell viability of human lung and skin cells exposed to different metal oxide nanomaterials

A quasi-QSAR model was developed to predict the cell viability of human lung (BEAS-2B) and skin (HaCaT) cells exposed to 21 types of metal oxide nanomaterials. A wide range of toxicity datasets obtained from the S2NANO (www.s2nano.org) database was used. The data of descriptors representing the physicochemical properties and experimental conditions were coded to quasi-SMILES. In particular, hierarchical cluster analysis (HCA) and min-max normalization method were respectively used in assigning alphanumeric codes for numerical descriptors (e.g., core size, hydrodynamic size, surface charge, and dose) and then quasi-QSAR model performances for both methods were compared. The quasi-QSAR models were developed using CORAL software (www.insilico.eu/coral). Quasi-QSAR model built using quasi-SMILES generated by means of HCA showed better performance than the min-max normalization method. The model showed satisfactory statistical results (R_adj² for the training dataset: 0.71-0.73; R_adj² for the calibration dataset: 0.74-0.82; and R_adj² for the validation dataset: 0.70-0.76).

Convolutional Embedding of Attributed Molecular Graphs for Physical Property Prediction

The task of learning an expressive molecular representation is central to developing quantitative structure-activity and property relationships. Traditional approaches rely on group additivity rules, empirical measurements or parameters, or generation of thousands of descriptors. In this paper, we employ a convolutional neural network for this embedding task by treating molecules as undirected graphs with attributed nodes and edges. Simple atom and bond attributes are used to construct atom-specific feature vectors that take into account the local chemical environment using different neighborhood radii. By working directly with the full molecular graph, there is a greater opportunity for models to identify important features relevant to a prediction task. Unlike other graph-based approaches, our atom featurization preserves molecule-level spatial information that significantly enhances model performance. Our models learn to identify important features of atom clusters for the prediction of aqueous solubility, octanol solubility, melting point, and toxicity. Extensions and limitations of this strategy are discussed.

QSPR/QSAR Analyses by Means of the CORAL Software

In this chapter, the methodology of building up quantitative structure—property/activity relationships (QSPRs/QSARs)—by means of the CORAL software is described. The Monte Carlo method is the basis of this approach. Simplified Molecular Input-Line Entry System (SMILES) is used as the representation of the molecular structure. The conversion of SMILES into the molecular graph is available for QSPR/QSAR analysis using the CORAL software. The model for an endpoint is a mathematical function of the correlation weights for various features of the molecular structure. Hybrid models that are based on features extracted from both SMILES and a graph also can be built up by the CORAL software. The conceptually new ideas collected and revealed through the CORAL software are: (1) any QSPR/QSAR model is a random event; and (2) optimal descriptor can be a translator of eclectic information into an endpoint prediction.

Use of the Monte Carlo Method for OECD Principles-Guided QSAR Modeling of SIRT1 Inhibitors

SIRT1 inhibitors offer therapeutic potential for the treatment of a number of diseases including cancer and human immunodeficiency virus infection. A diverse series of 45 compounds with reported SIRT1 inhibitory activity has been employed for the development of quantitative structure-activity relationship (QSAR) models using the Monte Carlo optimization method. This method makes use of simplified molecular input line entry system notation of the molecular structure. The QSAR models were built up according to OECD principles. Three subsets of three splits were examined and validated by respective external sets. All the three described models have good statistical quality. The best model has the following statistical characteristics: R²  = 0.8350, Q²_test  = 0.7491 for the test set and R²  = 0.9655, Q²_ext  = 0.9261 for the validation set. In the mechanistic interpretation, structural attributes responsible for the endpoint increase and decrease are defined. Further, the design of some prospective SIRT1 inhibitors is also presented on the basis of these structural attributes.

Optimal descriptor as a translator of eclectic data into prediction of cytotoxicity for metal oxide nanoparticles under different conditions

The Monte Carlo technique has been used to build up quantitative structure-activity relationships (QSARs) for prediction of dark cytotoxicity and photo-induced cytotoxicity of metal oxide nanoparticles to bacteria Escherichia coli (minus logarithm of lethal concentration for 50% bacteria pLC50, LC50 in mol/L). The representation of nanoparticles include (i) in the case of the dark cytotoxicity a simplified molecular input-line entry system (SMILES), and (ii) in the case of photo-induced cytotoxicity a SMILES plus symbol '^'. The predictability of the approach is checked up with six random distributions of available data into the visible training and calibration sets, and invisible validation set. The statistical characteristics of these models are correlation coefficient 0.90-0.94 (training set) and 0.73-0.98 (validation set).

QSPR/QSAR Analyses by Means of the CORAL Software

In this chapter, the methodology of building up quantitative structure—property/activity relationships (QSPRs/QSARs)—by means of the CORAL software is described. The Monte Carlo method is the basis of this approach. Simplified Molecular Input-Line Entry System (SMILES) is used as the representation of the molecular structure. The conversion of SMILES into the molecular graph is available for QSPR/QSAR analysis using the CORAL software. The model for an endpoint is a mathematical function of the correlation weights for various features of the molecular structure. Hybrid models that are based on features extracted from both SMILES and a graph also can be built up by the CORAL software. The conceptually new ideas collected and revealed through the CORAL software are: (1) any QSPR/QSAR model is a random event; and (2) optimal descriptor can be a translator of eclectic information into an endpoint prediction.

Applying machine learning techniques for ADME-Tox prediction: a review

INTRODUCTION

Pharmacokinetics involves the study of absorption, distribution, metabolism, excretion and toxicity of xenobiotics (ADME-Tox). In this sense, the ADME-Tox profile of a bioactive compound can impact its efficacy and safety. Moreover, efficacy and safety were considered some of the major causes of clinical failures in the development of new chemical entities. In this context, machine learning (ML) techniques have been often used in ADME-Tox studies due to the existence of compounds with known pharmacokinetic properties available for generating predictive models.

AREAS COVERED

This review examines the growth in the use of some ML techniques in ADME-Tox studies, in particular supervised and unsupervised techniques. Also, some critical points (e.g., size of the data set and type of output variable) must be considered during the generation of models that relate ADME-Tox properties and biological activity.

EXPERT OPINION

ML techniques have been successfully employed in pharmacokinetic studies, helping the complex process of designing new drug candidates from the use of reliable ML models. An application of this procedure would be the prediction of ADME-Tox properties from studies of quantitative structure-activity relationships or the discovery of new compounds from a virtual screening using filters based on results obtained from ML techniques.

Optimal descriptor as a translator of eclectic information into the prediction of membrane damage by means of various TiO(2) nanoparticles

The increasing use of nanomaterials incorporated into consumer products leads to the need for developing approaches to establish "quantitative structure-activity relationships" (QSARs) for various nanomaterials. However, the molecular structure as rule is not available for nanomaterials at least in its classic meaning. An possible alternative of classic QSAR (based on the molecular structure) is the using of data on physicochemical features of TiO(2) nanoparticles. The damage to cellular membranes (units L(-1)) by means of various TiO(2) nanoparticles is examined as the endpoint.

Validation of quantitative structure-activity relationship models to predict water-solubility of organic compounds

Water-solubility is an important physicochemical property in pharmaceutical and environmental studies. We assessed the performance of five predictive computer models: ACD/PhysChem History, ADMET Predictor, T.E.S.T., EPI Suite-WSKOWWIN and EPI Suite-WATERNT; two of them are commercial, the others are free. We used more than 4000 compounds with experimental values to evaluate the models, considering the chemicals inside and outside the applicability domain of the models, those used to build up the model (training set) and those not present in it (prediction set). We also evaluated their ability to predict continuous solubility values, and solubility classes. Overall, considering the whole data set, some models gave a good statistical performance, with R(2) up to 0.88.