Applicability Domains for Classification Problems: Benchmarking of Distance to Models for Ames Mutagenicity Set

Crowd computing: using competitive dynamics to develop and refine highly predictive models

A recent application of a crowd computing platform to develop highly predictive in silico models for use in the drug discovery process is described. The platform, Kaggle™, exploits a competitive dynamic that results in model optimization as the competition unfolds. Here, this dynamic is described in detail and compared with more-conventional modeling strategies. The complete and full structure of the underlying dataset is disclosed and some thoughts as to the broader utility of such 'gamification' approaches to the field of modeling are offered.

Comparison of in silico models for prediction of mutagenicity

Using a dataset with more than 6000 compounds, the performance of eight quantitative structure activity relationships (QSAR) models was evaluated: ACD/Tox Suite, Absorption, Distribution, Metabolism, Elimination, and Toxicity of chemical substances (ADMET) predictor, Derek, Toxicity Estimation Software Tool (T.E.S.T.), TOxicity Prediction by Komputer Assisted Technology (TOPKAT), Toxtree, CEASAR, and SARpy (SAR in python). In general, the results showed a high level of performance. To have a realistic estimate of the predictive ability, the results for chemicals inside and outside the training set for each model were considered. The effect of applicability domain tools (when available) on the prediction accuracy was also evaluated. The predictive tools included QSAR models, knowledge-based systems, and a combination of both methods. Models based on statistical QSAR methods gave better results.

QSAR Modelling of CYP3A4 Inhibition as a Screening Tool in the Context of DrugDrug Interaction Studies

Drugdrug interaction potential (DDI), especially cytochrome P450 (CYP) 3A4 inhibition potential, is one of the most important parameters to be optimized before preclinical and clinical pharmaceutical development as regard to the number of marketed drug metabolized mainly by this CYP and potentially co-administered with the future drug. The present study aims to develop in silico models for CYP3A4 inhibition prediction to help medicinal chemists during the discovery phase and even before the synthesis of new chemical entities (NCEs), focusing on NCEs devoid of any inhibitory potential toward this CYP. In order to find a relevant relationship between CYP3A4 inhibition and chemical features of the screened compounds, we applied a genetic-algorithm-based QSAR exploratory tool SQS (Stochastic QSAR Sampler) in combination with different description approaches comprising alignment-independent Volsurf descriptors, ISIDA fragments and Topological Fuzzy Pharmacophore Triplets. The experimental data used to build models were extracted from an in-house database. We derived a model with good prediction ability that was confirmed on both newly synthesized compound and public dataset retrieved from Pubchem database. This model is a promising efficient tool for filtering out potentially problematic compounds.

ToxAlerts: a Web server of structural alerts for toxic chemicals and compounds with potential adverse reactions

The article presents a Web-based platform for collecting and storing toxicological structural alerts from literature and for virtual screening of chemical libraries to flag potentially toxic chemicals and compounds that can cause adverse side effects. An alert is uniquely identified by a SMARTS template, a toxicological endpoint, and a publication where the alert was described. Additionally, the system allows storing complementary information such as name, comments, and mechanism of action, as well as other data. Most importantly, the platform can be easily used for fast virtual screening of large chemical datasets, focused libraries, or newly designed compounds against the toxicological alerts, providing a detailed profile of the chemicals grouped by structural alerts and endpoints. Such a facility can be used for decision making regarding whether a compound should be tested experimentally, validated with available QSAR models, or eliminated from consideration altogether. The alert-based screening can also be helpful for an easier interpretation of more complex QSAR models. The system is publicly accessible and tightly integrated with the Online Chemical Modeling Environment (OCHEM, http://ochem.eu). The system is open and expandable: any registered OCHEM user can introduce new alerts, browse, edit alerts introduced by other users, and virtually screen his/her data sets against all or selected alerts. The user sets being passed through the structural alerts can be used at OCHEM for other typical tasks: exporting in a wide variety of formats, development of QSAR models, additional filtering by other criteria, etc. The database already contains almost 600 structural alerts for such endpoints as mutagenicity, carcinogenicity, skin sensitization, compounds that undergo metabolic activation, and compounds that form reactive metabolites and, thus, can cause adverse reactions. The ToxAlerts platform is accessible on the Web at http://ochem.eu/alerts, and it is constantly growing.

Machine learning methods for property prediction in chemoinformatics: Quo Vadis?

This paper is focused on modern approaches to machine learning, most of which are as yet used infrequently or not at all in chemoinformatics. Machine learning methods are characterized in terms of the "modes of statistical inference" and "modeling levels" nomenclature and by considering different facets of the modeling with respect to input/ouput matching, data types, models duality, and models inference. Particular attention is paid to new approaches and concepts that may provide efficient solutions of common problems in chemoinformatics: improvement of predictive performance of structure-property (activity) models, generation of structures possessing desirable properties, model applicability domain, modeling of properties with functional endpoints (e.g., phase diagrams and dose-response curves), and accounting for multiple molecular species (e.g., conformers or tautomers).

Paradigm shift in toxicity testing and modeling

The limitations of traditional toxicity testing characterized by high-cost animal models with low-throughput readouts, inconsistent responses, ethical issues, and extrapolability to humans call for alternative strategies for chemical risk assessment. A new strategy using in vitro human cell-based assays has been designed to identify key toxicity pathways and molecular mechanisms leading to the prediction of an in vivo response. The emergence of quantitative high-throughput screening (qHTS) technology has proved to be an efficient way to decompose complex toxicological end points to specific pathways of targeted organs. In addition, qHTS has made a significant impact on computational toxicology in two aspects. First, the ease of mechanism of action identification brought about by in vitro assays has enhanced the simplicity and effectiveness of machine learning, and second, the high-throughput nature and high reproducibility of qHTS have greatly improved the data quality and increased the quantity of training datasets available for predictive model construction. In this review, the benefits of qHTS routinely used in the US Tox21 program will be highlighted. Quantitative structure-activity relationships models built on traditional in vivo data and new qHTS data will be compared and analyzed. In conjunction with the transition from the pilot phase to the production phase of the Tox21 program, more qHTS data will be made available that will enrich the data pool for predictive toxicology. It is perceivable that new in silico toxicity models based on high-quality qHTS data will achieve unprecedented reliability and robustness, thus becoming a valuable tool for risk assessment and drug discovery.

Development of in silico filters to predict activation of the pregnane X receptor (PXR) by structurally diverse drug-like molecules

The pregnane X receptor (PXR), a member of the nuclear hormone superfamily, regulates the expression of several enzymes and transporters involved in metabolically relevant processes. The significant induction of CYP450 enzymes by PXR, in particular CYP3A4, might significantly alter the metabolism of prescribed drugs. In order to early identify molecules in drug discovery with a potential to activate PXR as antitarget, we developed fast and reliable in silico filters by ligand-based QSAR techniques. Two classification models were established on a diverse dataset of 434 drug-like molecules. A second augmented set allowed focusing on interesting regions in chemical space. These classifiers are based on decision trees combined with a genetic algorithm based variable selection to arrive at predictive models. The classifier for the first dataset on 29 descriptors showed good performance on a test set with a correct classification of both 100% for PXR activators and non-activators plus 87% for activators and 83% for non-activators in an external dataset. The second classifier then correctly predicts 97% activators and 91% non-activators in a test set and 94% for activators and 64% non-activators in an external set of 50 molecules, which still qualifies for application as a filter focusing on PXR activators. Finally a quantitative model for PXR activation for a subset of these molecules was derived using a regression-tree approach combined with GA variable selection. This final model shows a predictive r(2) of 0.774 for the test set and 0.452 for an external set of 33 molecules. Thus, the combination of these filters consistently provide guidelines for lowering PXR activation in novel candidate molecules.

In silico assessment of chemical biodegradability

Biodegradation is the principal environmental dissipation process. Due to a lack of comprehensive experimental data, high study cost and time-consuming, in silico approaches for assessing the biodegradable profiles of chemicals are encouraged and is an active current research topic. Here we developed in silico methods to estimate chemical biodegradability in the environment. At first 1440 diverse compounds tested under the Japanese Ministry of International Trade and Industry (MITI) protocol were used. Four different methods, namely support vector machine, k-nearest neighbor, naïve Bayes, and C4.5 decision tree, were used to build the combinatorial classification probability models of ready versus not ready biodegradability using physicochemical descriptors and fingerprints separately. The overall predictive accuracies of the best models were more than 80% for the external test set of 164 diverse compounds. Some privileged substructures were further identified for ready or not ready biodegradable chemicals by combining information gain and substructure fragment analysis. Moreover, 27 new predicted chemicals were selected for experimental assay through the Japanese MITI test protocols, which validated that all 27 compounds were predicted correctly. The predictive accuracies of our models outperform the commonly used software of the EPI Suite. Our study provided critical tools for early assessment of biodegradability of new organic chemicals in environmental hazard assessment.

An investigation into pharmaceutically relevant mutagenicity data and the influence on Ames predictive potential

Background

In drug discovery, a positive Ames test for bacterial mutation presents a significant hurdle to advancing a drug to clinical trials. In a previous paper, we discussed success in predicting the genotoxicity of reagent-sized aryl-amines (ArNH₂), a structure frequently found in marketed drugs and in drug discovery, using quantum mechanics calculations of the energy required to generate the DNA-reactive nitrenium intermediate (ArNH:+). In this paper we approach the question of what molecular descriptors could improve these predictions and whether external data sets are appropriate for further training.

Results

In trying to extend and improve this model beyond this quantum mechanical reaction energy, we faced considerable difficulty, which was surprising considering the long history and success of QSAR model development for this test. Other quantum mechanics descriptors were compared to this reaction energy including AM1 semi-empirical orbital energies, nitrenium formation with alternative leaving groups, nitrenium charge, and aryl-amine anion formation energy. Nitrenium formation energy, regardless of the starting species, was found to be the most useful single descriptor. External sets used in other QSAR investigations did not present the same difficulty using the same methods and descriptors. When considering all substructures rather than just aryl-amines, we also noted a significantly lower performance for the Novartis set. The performance gap between Novartis and external sets persists across different descriptors and learning methods. The profiles of the Novartis and external data are significantly different both in aryl-amines and considering all substructures. The Novartis and external data sets are easily separated in an unsupervised clustering using chemical fingerprints. The chemical differences are discussed and visualized using Kohonen Self-Organizing Maps trained on chemical fingerprints, mutagenic substructure prevalence, and molecular weight.

Conclusions

Despite extensive work in the area of predicting this particular toxicity, work in designing and publishing more relevant test sets for compounds relevant to drug discovery is still necessary. This work also shows that great care must be taken in using QSAR models to replace experimental evidence. When considering all substructures, a random forest model, which can inherently cover distinct neighborhoods, built on Novartis data and previously reported external data provided a suitable model.

The acceptance of in silico models for REACH: Requirements, barriers, and perspectives

In silico models have prompted considerable interest and debate because of their potential value in predicting the properties of chemical substances for regulatory purposes. The European REACH legislation promotes innovation and encourages the use of alternative methods, but in practice the use of in silico models is still very limited. There are many stakeholders influencing the regulatory trajectory of quantitative structure-activity relationships (QSAR) models, including regulators, industry, model developers and consultants. Here we outline some of the issues and challenges involved in the acceptance of these methods for regulatory purposes.

Avoidance of the Ames test liability for aryl-amines via computation

Aryl-amines are commonly used synthons in modern drug discovery, however a minority of these chemical templates have the potential to cause toxicity through mutagenicity. The toxicity mostly arises through a series of metabolic steps leading to a reactive electrophilic nitrenium cation intermediate that reacts with DNA nucleotides causing mutation. Highly detailed in silico calculations of the energetics of chemical reactions involved in the metabolic formation of nitrenium cations have been performed. This allowed a critical assessment of the accuracy and reliability of using a theoretical formation energy of the DNA-reactive nitrenium intermediate to correlate with the Ames test response. This study contains the largest data set reported to date, and presents the in silico calculations versus the in vitro Ames response data in the form of beanplots commonly used in statistical analysis. A comparison of this quantum mechanical approach to QSAR and knowledge-based methods is also reported, as well as the calculated formation energies of nitrenium ions for thousands of commercially available aryl-amines generated as a watch-list for medicinal chemists in their synthetic optimization strategies.