AI and SAR approaches for predicting chemical carcinogenicity: survey and status report

Molecular docking of antidiabetic molecules of libas (Spondias pinnata) fruit and prediction of their pharmacokinetic properties

Diabetes mellitus is one of the chronic metabolic disorders that affects more than 16 million Filipinos. Proper education, medical intervention, and a good lifestyle can help individuals control and manage this disease. Spondias pinnata is one of the underutilized crops in the Philippines that is well-known for its satisfactory flavor and medicinal properties, including its antidiabetic activity. The quest for a natural and effective drug to manage diseases is a continuous work in progress. Drug discovery and design is a tedious and expensive process. Computer-aided drug design guides the design and makes the process more efficient and less costly. Molecular docking was used to determine the potential antidiabetic compounds from the 48 reported compounds found in S. pinnata fruit. Seven compounds namely squalene (-9.1 kcal/mol), rutin (-9 kcal/mol), catechin (-8.7 kcal/mol), quercetin (-8.5 kcal/mol), tocopherol (-8.4 kcal/mol), myricetin (-8.4 kcal/mol), and ellagic acid (-8.3 kcal/mol) showed binding affinities comparable to those of pioglitazone, a standard drug, with peroxisome proliferator-activated receptor gamma (PPARγ). Tocopherol and catechin showed good ADMET properties. Among the two compounds, catechin passed the four filters for drug-likeness. Thus, catechin could be a potential compound for the development of antidiabetic drugs.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40203-024-00230-3.

Combination of high resolution mass spectrometry and a halogen extraction code to identify chlorinated disinfection byproducts formed from aromatic amino acids

Chlorination can lead to the formation of hazardous chlorinated disinfection byproducts (Cl-DBPs). We identified tyrosine (Tyr) and tryptophan (Trp) as precursors of toxic Cl-DBPs and developed a halogen extraction code to complement ultra performance liquid chromatography in tandem with high resolution mass spectrometry (UPLC-HRMS) in detecting and identifying Cl-DBPs. We detected 20 and 11 Cl-DBPs formed from chlorination of Tyr and Trp, respectively, and identified the structures of 15 Cl-DBPs. Fourteen structures were previously unreported. We also proposed the tentative formation pathways of these newly identified Cl-DBPs. Their incidence in real water sources demonstrated that these Cl-DBPs are likely to form during chlorination of reclaimed water. We computationally predicted the toxicity of these Cl-DBPs, which was relatively high, indicating that these Cl-DBPs could be hazardous and were of valid concern. Combining analytical data with the halogen extraction code can identify Cl-DBPs accurately from complex compounds. This analytical method can be used to identify Cl-DBPs of water treatment procedures in further studies.

Quantitative structure-activity relationship models for compounds with anticonvulsant activity

Introduction: Third-generation antiepileptic drugs have seemingly failed to improve the global figures of seizure control and can still be regarded as symptomatic treatments. Quantitative structure-activity relationships (QSAR) can be used to guide hit-to-lead and lead optimization projects and applied to the large-scale virtual screening of chemical libraries. Areas covered: In this review, the authors cover reports on QSAR models related to antiepileptic drugs and drug targets in epilepsy, analyzing whether they refer to classic or non-classic QSAR and if they apply QSAR as a descriptive or predictive approach, among other considerations. The article finally focuses on a more detailed discussion of those predictive studies which include some sort of experimental validation, i.e. papers in which the reported models have been used to identify novel active compounds which have been tested in vitro and/or in vivo. Expert opinion: There are significant opportunities to apply the QSAR methodology to assist the discovery of more efficacious antiepileptic drugs. Considering the intrinsic complexity of the disorder, such applications should focus on state-of-the-art approximations (e.g. systemic, multi-target and multi-scale QSAR as well as ensemble and deep learning) and modeling the effects on novel drug targets and modern screening tools.

The application of molecular topology for ulcerative colitis drug discovery

INTRODUCTION

Although the therapeutic arsenal against ulcerative colitis has greatly expanded (including the revolutionary advent of biologics), there remain patients who are refractory to current medications while the safety of the available therapeutics could also be improved. Molecular topology provides a theoretic framework for the discovery of new therapeutic agents in a very efficient manner, and its applications in the field of ulcerative colitis have slowly begun to flourish. Areas covered: After discussing the basics of molecular topology, the authors review QSAR models focusing on validated targets for the treatment of ulcerative colitis, entirely or partially based on topological descriptors. Expert opinion: The application of molecular topology to ulcerative colitis drug discovery is still very limited, and many of the existing reports seem to be strictly theoretic, with no experimental validation or practical applications. Interestingly, mechanism-independent models based on phenotypic responses have recently been reported. Such models are in agreement with the recent interest raised by network pharmacology as a potential solution for complex disorders. These and other similar studies applying molecular topology suggest that some therapeutic categories may present a 'topological pattern' that goes beyond a specific mechanism of action.

Grouping of CoMPARA data with respect to compounds from the carcinogenic potency database

Methods for clustering and measures of similarity of chemical structures have become an important supporting tool in chemoinformatics. They represent the basis for categorization of chemicals and read-across, where a molecular property is estimated from 'similar molecules'. This study proposes a clustering scheme within the given dataset with respect to a reference dataset. The scheme was applied on two datasets ToxCast_AR_Agonist and ToxCast_AR_Antagonists with 1654 and 1522 compounds, respectively. The compounds are tested to androgen receptor activity (AR) in 11 high throughput screening assays. The carcinogenic dataset was used as the reference set.

Novel Uses of In Vitro Data to Develop Quantitative Biological Activity Relationship Models for in Vivo Carcinogenicity Prediction

The availability of large in vitro datasets enables better insight into the mode of action of chemicals and better identification of potential mechanism(s) of toxicity. Several studies have shown that not all in vitro assays can contribute as equal predictors of in vivo carcinogenicity for development of hybrid Quantitative Structure Activity Relationship (QSAR) models. We propose two novel approaches for the use of mechanistically relevant in vitro assay data in the identification of relevant biological descriptors and development of Quantitative Biological Activity Relationship (QBAR) models for carcinogenicity prediction. We demonstrate that in vitro assay data can be used to develop QBAR models for in vivo carcinogenicity prediction via two case studies corroborated with firm scientific rationale. The case studies demonstrate the similarities between QBAR and QSAR modeling in: (i) the selection of relevant descriptors to be used in the machine learning algorithm, and (ii) the development of a computational model that maps chemical or biological descriptors to a toxic endpoint. The results of both the case studies show: (i) improved accuracy and sensitivity which is especially desirable under regulatory requirements, and (ii) overall adherence with the OECD/REACH guidelines. Such mechanism based models can be used along with QSAR models for prediction of mechanistically complex toxic endpoints.

Prediction of acute mammalian toxicity using QSAR methods: a case study of sulfur mustard and its breakdown products

Predicting toxicity quantitatively, using Quantitative Structure Activity Relationships (QSAR), has matured over recent years to the point that the predictions can be used to help identify missing comparison values in a substance's database. In this manuscript we investigate using the lethal dose that kills fifty percent of a test population (LD₅₀) for determining relative toxicity of a number of substances. In general, the smaller the LD₅₀ value, the more toxic the chemical, and the larger the LD₅₀ value, the lower the toxicity. When systemic toxicity and other specific toxicity data are unavailable for the chemical(s) of interest, during emergency responses, LD₅₀ values may be employed to determine the relative toxicity of a series of chemicals. In the present study, a group of chemical warfare agents and their breakdown products have been evaluated using four available rat oral QSAR LD₅₀ models. The QSAR analysis shows that the breakdown products of Sulfur Mustard (HD) are predicted to be less toxic than the parent compound as well as other known breakdown products that have known toxicities. The QSAR estimated break down products LD₅₀ values ranged from 299 mg/kg to 5,764 mg/kg. This evaluation allows for the ranking and toxicity estimation of compounds for which little toxicity information existed; thus leading to better risk decision making in the field.

First report on development of quantitative interspecies structure-carcinogenicity relationship models and exploring discriminatory features for rodent carcinogenicity of diverse organic chemicals using OECD guidelines

Different regulatory agencies in food and drug administration and environmental protection worldwide are employing quantitative structure-activity relationship (QSAR) models to fill the data gaps related with properties of chemicals affecting the environment and human health. Carcinogenicity is a toxicity endpoint of major concern in recent times. Interspecies toxicity correlations may provide a tool for estimating sensitivity towards toxic chemical exposure with known levels of uncertainty for a diversity of wildlife species. In this background, we have developed quantitative interspecies structure-carcinogenicity correlation models for rat and mouse [rodent species according to the Organization for Economic Cooperation and Development (OECD) guidelines] based on the carcinogenic potential of 166 organic chemicals with wide diversity of molecular structures, spanning a large number of chemical classes and biological mechanisms. All the developed models have been assessed according to the OECD principles for the validation of QSAR models. Consensus predictions for carcinogenicity of the individual compounds are presented here for any one species when the data for the other species are available. Informative illustrations of the contributing structural fragments of chemicals which are responsible for specific carcinogenicity endpoints are identified by the developed models. The models have also been used to predict mouse carcinogenicities of 247 organic chemicals (for which rat carcinogenicities are present) and rat carcinogenicities of 150 chemicals (for which mouse carcinogenicities are present). Discriminatory features for rat and mouse carcinogenicity values have also been explored.

Expert Systems for Toxicity Prediction

Expert systems offer the facility to predict a toxicity endpoint, as well sometimes as additional relevant information, simply by inputting the chemical structure of a compound. There is now a number of expert systems available, mostly on a commercial basis although a few are free to use or download. This chapter discusses nineteen currently available expert systems, and their performances (if known). Published studies of consensus predictions with these expert systems indicate that these give better results than do individual expert systems.

A test set of compounds with Tetrahymena pyriformis toxicities has been run through the two expert systems known to predict these toxicities; the predictions were quite good, with standard errors of prediction of 0.395 and 0.433 log unit. A further test set of compounds with local lymph node assay skin sensitisation data has been run through seven expert systems, and it was found that consensus predictions were better than were those from any individual expert system.

Evaluation of the OECD (Q)SAR Application Toolbox and Toxtree for predicting and profiling the carcinogenic potential of chemicals

The OECD (Q)SAR Application Toolbox and Toxtree are software tools used in regulatory toxicology to fill gaps in (eco)toxicity data. They include different SAR and QSAR models for estimating (eco)toxicological endpoints. Among them, the Benigni/Bossa rule-based system is proposed to characterize the carcinogenic potential of chemicals. Our study evaluates the predictive performance that can be expected from the OECD (Q)SAR Toolbox and Toxtree when analysing chemicals by means of the structural alerts coded within the Benigni/Bossa rule-based system for carcinogenicity and the associated QSAR model (QSAR8). These evaluations have been carried out thanks to a large collection of chemicals retrieved from original publications and public databases. Overall, our findings confirm the performance of the system of structural alerts while suggesting that the sensitivity of QSAR8, as implemented in the two tools, is lower than what was previously reported. They also indicate that attention has to be paid when interpreting the output of the two tools because of possible malfunctions involving the coding of two-dimensional structures. A set of possible modulating factors for the structural alert identifying polycyclic aromatic hydrocarbons is also proposed together with candidates for putative new structural alerts not included in the tested tools.

New public QSAR model for carcinogenicity

BACKGROUND

One of the main goals of the new chemical regulation REACH (Registration, Evaluation and Authorization of Chemicals) is to fulfill the gaps in data concerned with properties of chemicals affecting the human health. (Q)SAR models are accepted as a suitable source of information. The EU funded CAESAR project aimed to develop models for prediction of 5 endpoints for regulatory purposes. Carcinogenicity is one of the endpoints under consideration.

RESULTS

Models for prediction of carcinogenic potency according to specific requirements of Chemical regulation were developed. The dataset of 805 non-congeneric chemicals extracted from Carcinogenic Potency Database (CPDBAS) was used. Counter Propagation Artificial Neural Network (CP ANN) algorithm was implemented. In the article two alternative models for prediction carcinogenicity are described. The first model employed eight MDL descriptors (model A) and the second one twelve Dragon descriptors (model B). CAESAR's models have been assessed according to the OECD principles for the validation of QSAR. For the model validity we used a wide series of statistical checks. Models A and B yielded accuracy of training set (644 compounds) equal to 91% and 89% correspondingly; the accuracy of the test set (161 compounds) was 73% and 69%, while the specificity was 69% and 61%, respectively. Sensitivity in both cases was equal to 75%. The accuracy of the leave 20% out cross validation for the training set of models A and B was equal to 66% and 62% respectively. To verify if the models perform correctly on new compounds the external validation was carried out. The external test set was composed of 738 compounds. We obtained accuracy of external validation equal to 61.4% and 60.0%, sensitivity 64.0% and 61.8% and specificity equal to 58.9% and 58.4% respectively for models A and B.

CONCLUSION

Carcinogenicity is a particularly important endpoint and it is expected that QSAR models will not replace the human experts opinions and conventional methods. However, we believe that combination of several methods will provide useful support to the overall evaluation of carcinogenicity. In present paper models for classification of carcinogenic compounds using MDL and Dragon descriptors were developed. Models could be used to set priorities among chemicals for further testing. The models at the CAESAR site were implemented in java and are publicly accessible.

New public QSAR model for carcinogenicity

BACKGROUND

One of the main goals of the new chemical regulation REACH (Registration, Evaluation and Authorization of Chemicals) is to fulfill the gaps in data concerned with properties of chemicals affecting the human health. (Q)SAR models are accepted as a suitable source of information. The EU funded CAESAR project aimed to develop models for prediction of 5 endpoints for regulatory purposes. Carcinogenicity is one of the endpoints under consideration.

RESULTS

Models for prediction of carcinogenic potency according to specific requirements of Chemical regulation were developed. The dataset of 805 non-congeneric chemicals extracted from Carcinogenic Potency Database (CPDBAS) was used. Counter Propagation Artificial Neural Network (CP ANN) algorithm was implemented. In the article two alternative models for prediction carcinogenicity are described. The first model employed eight MDL descriptors (model A) and the second one twelve Dragon descriptors (model B). CAESAR's models have been assessed according to the OECD principles for the validation of QSAR. For the model validity we used a wide series of statistical checks. Models A and B yielded accuracy of training set (644 compounds) equal to 91% and 89% correspondingly; the accuracy of the test set (161 compounds) was 73% and 69%, while the specificity was 69% and 61%, respectively. Sensitivity in both cases was equal to 75%. The accuracy of the leave 20% out cross validation for the training set of models A and B was equal to 66% and 62% respectively. To verify if the models perform correctly on new compounds the external validation was carried out. The external test set was composed of 738 compounds. We obtained accuracy of external validation equal to 61.4% and 60.0%, sensitivity 64.0% and 61.8% and specificity equal to 58.9% and 58.4% respectively for models A and B.

CONCLUSION

Carcinogenicity is a particularly important endpoint and it is expected that QSAR models will not replace the human experts opinions and conventional methods. However, we believe that combination of several methods will provide useful support to the overall evaluation of carcinogenicity. In present paper models for classification of carcinogenic compounds using MDL and Dragon descriptors were developed. Models could be used to set priorities among chemicals for further testing. The models at the CAESAR site were implemented in java and are publicly accessible.

Counter propagation artificial neural network categorical models for prediction of carcinogenicity for non-congeneric chemicals

One of the main goals of the new chemical regulation REACH (Registration, Evaluation and Authorization of Chemicals) is to fill the gaps on the toxicological properties of chemicals that affect human health. Carcinogenicity is one of the endpoints under consideration. The information obtained from (quantitative) structure-activity relationship ((Q)SAR) models is accepted as an alternative solution to avoid expensive and time-consuming animal tests. The reported results were obtained within the framework of the European project 'Computer Assisted Evaluation of industrial chemical Substances According to Regulations (CAESAR)'. In this article, we demonstrate intermediate results for counter propagation artificial neural network (CP ANN) models for the prediction category of the carcinogenic potency using two-dimensional (2D) descriptors from different software programs. A total of 805 non-congeneric chemicals were extracted from the Carcinogenic Potency Database (CPDBAS). The resulting models had prediction accuracies for internal (training) and external (test) sets as high as 91-93% and 68-70%, respectively. The sensitivity and specificity of the test set were 69-73 and 63-72% correspondingly. High specificity is critical in models for regulatory use that are aimed at ensuring public safety. Thus, the errors that give rise to false negatives are much more relevant. We discuss how we can increase the number of correctly predicted carcinogens using the correlation between the threshold and the values of the sensitivity and specificity.

Quantitative structure-activity relationship modeling of rat acute toxicity by oral exposure

Few quantitative structure-activity relationship (QSAR) studies have successfully modeled large, diverse rodent toxicity end points. In this study, a comprehensive data set of 7385 compounds with their most conservative lethal dose (LD(50)) values has been compiled. A combinatorial QSAR approach has been employed to develop robust and predictive models of acute toxicity in rats caused by oral exposure to chemicals. To enable fair comparison between the predictive power of models generated in this study versus a commercial toxicity predictor, TOPKAT (Toxicity Prediction by Komputer Assisted Technology), a modeling subset of the entire data set was selected that included all 3472 compounds used in TOPKAT's training set. The remaining 3913 compounds, which were not present in the TOPKAT training set, were used as the external validation set. QSAR models of five different types were developed for the modeling set. The prediction accuracy for the external validation set was estimated by determination coefficient R(2) of linear regression between actual and predicted LD(50) values. The use of the applicability domain threshold implemented in most models generally improved the external prediction accuracy but expectedly led to the decrease in chemical space coverage; depending on the applicability domain threshold, R(2) ranged from 0.24 to 0.70. Ultimately, several consensus models were developed by averaging the predicted LD(50) for every compound using all five models. The consensus models afforded higher prediction accuracy for the external validation data set with the higher coverage as compared to individual constituent models. The validated consensus LD(50) models developed in this study can be used as reliable computational predictors of in vivo acute toxicity.

Integrated decision-tree testing strategies for mutagenicity and carcinogenicity with respect to the requirements of the EU REACH legislation

Liverpool John Moores University and FRAME recently conducted a research project sponsored by Defra, on the status of alternatives to animal testing with regard to the European Union REACH (Registration, Evaluation and Authorisation of Chemicals) system for the safety testing and risk assessment of chemicals. The project covered all the main toxicity endpoints associated with the REACH system. This paper focuses on the prospects for using alternative methods (both in vitro and in silico) for mutagenicity (genotoxicity) and carcinogenicity testing--two toxicity endpoints, which, together with reproductive toxicity, are of pivotal importance for the REACH system. The manuscript critically discusses well-established testing approaches, and in particular, the requirement for short-term in vivo tests for confirming positive mutagenicity, and the need for the rodent bioassay for detecting non-genotoxic carcinogens. Recently-proposed testing strategies focusing on non-animal approaches are also considered, and our own testing scheme is presented and supported with background information. This scheme makes maximum use of pre-existing data, computer (in silico) and in vitro methods, with weight-of-evidence assessments at each major stage. The need for the improvement of in vitro methods, to reduce the generation of false-positive results, is also discussed. Lastly, ways in which reduction and refinement measures can be used are also considered, and some recommendations are made for future research to facilitate the implementation of the proposed testing scheme.

Combinatorial QSAR modeling of chemical toxicants tested against Tetrahymena pyriformis

Selecting most rigorous quantitative structure-activity relationship (QSAR) approaches is of great importance in the development of robust and predictive models of chemical toxicity. To address this issue in a systematic way, we have formed an international virtual collaboratory consisting of six independent groups with shared interests in computational chemical toxicology. We have compiled an aqueous toxicity data set containing 983 unique compounds tested in the same laboratory over a decade against Tetrahymena pyriformis. A modeling set including 644 compounds was selected randomly from the original set and distributed to all groups that used their own QSAR tools for model development. The remaining 339 compounds in the original set (external set I) as well as 110 additional compounds (external set II) published recently by the same laboratory (after this computational study was already in progress) were used as two independent validation sets to assess the external predictive power of individual models. In total, our virtual collaboratory has developed 15 different types of QSAR models of aquatic toxicity for the training set. The internal prediction accuracy for the modeling set ranged from 0.76 to 0.93 as measured by the leave-one-out cross-validation correlation coefficient ( Q abs2). The prediction accuracy for the external validation sets I and II ranged from 0.71 to 0.85 (linear regression coefficient R absI2) and from 0.38 to 0.83 (linear regression coefficient R absII2), respectively. The use of an applicability domain threshold implemented in most models generally improved the external prediction accuracy but at the same time led to a decrease in chemical space coverage. Finally, several consensus models were developed by averaging the predicted aquatic toxicity for every compound using all 15 models, with or without taking into account their respective applicability domains. We find that consensus models afford higher prediction accuracy for the external validation data sets with the highest space coverage as compared to individual constituent models. Our studies prove the power of a collaborative and consensual approach to QSAR model development. The best validated models of aquatic toxicity developed by our collaboratory (both individual and consensus) can be used as reliable computational predictors of aquatic toxicity and are available from any of the participating laboratories.

Proposed integrated decision-tree testing strategies for mutagenicity and carcinogenicity in relation to the EU REACH legislation

Liverpool John Moores University and FRAME recently conducted a research project sponsored by Defra, on the status of alternatives to animal testing with regard to the European Union REACH (Registration, Evaluation and Authorisation of Chemicals) system for the safety testing and risk assessment of chemicals. The project covered all the main toxicity endpoints associated with the REACH system. This paper focuses on the prospects for using alternative methods (both in vitro and in silico) for mutagenicity (genotoxicity) and carcinogenicity testing - two toxicity endpoints, which, together with reproductive toxicity, are of pivotal importance for the REACH system. The manuscript critically discusses well-established testing approaches, and in particular, the requirement for short-term in vivo tests for confirming positive mutagenicity, and the need for the rodent bioassay for detecting non-genotoxic carcinogens. Recently-proposed testing strategies focusing on non-animal approaches are also considered, and our own testing scheme is presented and supported with background information. This scheme makes maximum use of pre-existing data, computer (in silico) and in vitro methods, with weight-of-evidence assessments at each major stage. The need for the improvement of in vitro methods, to reduce the generation of false-positive results, is also discussed. Lastly, ways in which reduction and refinement measures can be used are also considered, and some recommendations are made for future research to facilitate the implementation of the proposed testing scheme.

Facilitating compound progression of antiretroviral agents via modeling and simulation

Pharmacotherapy in human immunodeficiency virus (HIV)-infected patients and the development of safe and effective antiretroviral dosing regimens has been hindered by numerous issues, including the rapid development of viral resistance to drug therapy, the narrow therapeutic window of the drug compounds, and lack of fundamental knowledge concerning the sources of variation in exposure and response to antiretroviral agents. Sources of variation may include factors such as interpatient differences in genetic expression, immunological response, pathogenesis, epidemiologic and socioeconomic factors, and demographics. Modeling and simulation (M&S) techniques have become valuable tools to identify and quantify variability in exposure and response to antiretroviral agents throughout the drug development process. Before actual entry into human safety and pharmacokinetic (PK) trials, in vitro screening and in vivo pharmacology studies conducted to assess compound potency and compatibility with agents included in acceptable antiretroviral therapy (ART) regimens can be characterized via quantitative relationships. In addition, physiochemical data is initially used to screen drug candidates based on favorable PK and biopharmaceutic properties. Compound progression can likewise be supported with M&S exercises to ensure the traceability of key assumptions and decisions. The underlying techniques utilize nonlinear mixed effect modeling, Monte Carlo simulation, Neural networks, several regression-based approaches, and less computationally intensive techniques. The application of such an approach promises to be an essential component in the development of new agents to treat HIV-1 and is being implemented in the context of evaluating Nk1r antagonists as potential candidates to treat NeuroAIDS.

PASS: identification of probable targets and mechanisms of toxicity

Toxicity of chemical compound is a complex phenomenon that may be caused by its interaction with different targets in the organism. Two distinct types of toxicity can be broadly specified: the first one is caused by the strong compound's interaction with a single target (e.g. AChE inhibition); while the second one is caused by the moderate compound's interaction with many various targets. Computer program PASS predicts about 2500 kinds of biological activities based on the structural formula of chemical compounds. Prediction is based on the robust analysis of structure-activity relationships for about 60,000 biologically active compounds. Mean accuracy exceeds 90% in leave-one-out cross-validation. In addition to some kinds of adverse effects and specific toxicity (e.g. carcinogenicity, mutagenicity, etc.), PASS predicts approximately 2000 kinds of biological activities at the molecular level, that providing an estimated profile of compound's action in biological space. Such profiles can be used to recognize the most probable targets, interaction with which might be a reason of compound's toxicity. Applications of PASS predictions for analysis of probable targets and mechanisms of toxicity are discussed.

Use of computer-assisted prediction of toxic effects of chemical substances

The current revision of the European policy for the evaluation of chemicals (REACH) has lead to a controversy with regard to the need of additional animal safety testing. To avoid increases in animal testing but also to save time and resources, alternative in silico or in vitro tests for the assessment of toxic effects of chemicals are advocated. The draft of the original document issued in 29th October 2003 by the European Commission foresees the use of alternative methods but does not give further specification on which methods should be used. Computer-assisted prediction models, so-called predictive tools, besides in vitro models, will likely play an essential role in the proposed repertoire of "alternative methods". The current discussion has urged the Advisory Committee of the German Toxicology Society to present its position on the use of predictive tools in toxicology. Acceptable prediction models already exist for those toxicological endpoints which are based on well-understood mechanism, such as mutagenicity and skin sensitization, whereas mechanistically more complex endpoints such as acute, chronic or organ toxicities currently cannot be satisfactorily predicted. A potential strategy to assess such complex toxicities will lie in their dissection into models for the different steps or pathways leading to the final endpoint. Integration of these models should result in a higher predictivity. Despite these limitations, computer-assisted prediction tools already today play a complementary role for the assessment of chemicals for which no data is available or for which toxicological testing is impractical due to the lack of availability of sufficient compounds for testing. Furthermore, predictive tools offer support in the screening and the subsequent prioritization of compound for further toxicological testing, as expected within the scope of the European REACH program. This program will also lead to the collection of high-quality data which will broaden the database for further (Q)SAR approaches and will in turn increase the predictivity of predictive tools.

Prediction of the rodent carcinogenicity of 60 pesticides by the DEREKfW expert system

The two-year rodent bioassay represents the golden standard for evaluating the carcinogenicity of chemicals. Because of practical and ethical reasons, alternative approaches have been investigated for many years. Among these approaches, the (quantitative) structure-activity relationships [(Q)SARs] offer promising perspectives for quickly screening a large number of chemicals. To increase the acceptance of (Q)SARs among the regulators, their predictive power needs to be scientifically validated. In this article, we tested the capacity of the DEREKfW expert system to qualitatively predict the rodent carcinogenicity and the genotoxic potential of 60 pesticides recently registered in Switzerland. The percentage of false negatives was found to be 31% for carcinogenicity. The associated sensitivity of 69% indicates that most of the pesticides with positive rodent bioassay results were detected by DEREKfW. On the other hand, the low specificity of 47% indicates that many pesticides may be flagged as carcinogenic while rodent bioassays would not confirm this potential. This may lead to unnecessary testing or the unnecessary restriction of a chemical.

A radial-distribution-function approach for predicting rodent carcinogenicity

Carcinogenic activity has been investigated using the Radial-Distribution-Function (RDF) approach. A discriminant model was developed to predict the carcinogenic and non-carcinogenic activity on a data set of 188 compounds. The percentage of overall classification was 76.4% for the carcinogenic chemicals and 72.5% for the non-carcinogenic chemicals. The predictive power of the model was validated by two tests: a cross-validation by the resubstitution technique and a test set (compounds not used in the development of the model) with 79.3 and 72.5% good classification, respectively. The RDF descriptors were compared with eight other methodologies; Constitutional, Molecular walks counts, Galvez topological charge indices, 2D autocorrelations, Randić molecular profiles, Geometrical, 3D-MORSE, and WHIM, demonstrating that the RDF descriptors are better to the rest of the approaches used.

Computer-aided rodent carcinogenicity prediction

The potential of the computer program PASS (Prediction Activity Spectra for Substances) to predict rodent carcinogenicity for chemical compounds was studied. PASS predicts carcinogenicity of chemical compounds on the basis of their structural formula and of structure-activity relationship analysis of known carcinogens and non-carcinogens. The data on structures and experimental results of 2-year carcinogenicity assays for 412 chemicals from the NTP (National Toxicological Program) and 1190 chemicals from the CPDB (Carcinogenic Potency Database) were used in our study. The predictions take into consideration information about species and sex of animals. For evaluation of the predictive accuracy we used two procedures: leave-one-out cross-validation (LOO CV) and leave-20%-out cross-validation. In the last case we randomly divided the studied data set 20 times into two subsets. The data from the first subset, containing 80% of the compounds, were added to the PASS training set (which includes about 46,000 compounds with about 1500 biological activity types collected during the last 20 years to predict biological activity spectra), the second subset with 20% of the compounds was used as an evaluation set. The mean accuracy of prediction calculated by LOO CV is about 73% for NTP compounds in the 'equivocal' category of carcinogenic activity and 80% for NTP compounds in the 'evidence' category of carcinogenicity. The mean accuracy of prediction for the CPDB database is 89.9% calculated by LOO CV and 63.4% calculated by leave-20%-out cross-validation. Influence of incorporation of species and sex data on the accuracy of carcinogenicity prediction was also investigated. It was shown that the accuracy was increased only for data on male animals.

A topological substructural approach applied to the computational prediction of rodent carcinogenicity

The carcinogenic activity has been investigated by using a topological substructural molecular design approach (TOPS-MODE). A discriminant model was developed to predict the carcinogenic and noncarcinogenic activity on a data set of 189 compounds. The percentage of correct classification was 76.32%. The predictive power of the model was validated by three test: an external test set (compounds not used in the develop of the model, with a 72.97% of good classification), a leave-group-out cross-validation procedure (4-fold full cross-validation, removing 20% of compounds in each cycle, with a good prediction of 76.31%) and two external prediction sets (the first and second exercises of the National Toxicology Program). This methodology evidenced that the hydrophobicity increase the carcinogenic activity and the dipole moment of the molecule decrease it; suggesting the capacity of the TOPS-MODE descriptors to estimate this property for new drug candidates. Finally, the positive and negative fragment contributions to the carcinogenic activity were identified (structural alerts) and their potentialities in the lead generation process and in the design of 'safer' chemicals were evaluated.

Challenges in constructing statistically based structure-activity relationship models for developmental toxicity

Regulatory agencies are increasingly called upon to review large numbers of environmental contaminants that have not been characterized for their potential to pose a health risk. Additionally, there is special interest in protecting potentially sensitive subpopulations and identifying developmental toxicants that may be present in the environment. Thus, there is an urgent need for efficient methods to screen large numbers of chemicals for their potential to pose a developmental hazard. One potential screening method involves the use of statistically based structure-activity relationship (SAR) tools to predict activity of untested chemicals. Such systems rely on statistical analyses to discern relationships between structure and activity for a training set of substances. Predictions can then be made for an untested substance as long as its structural features are encompassed by chemicals of the training set. In theory, such systems could assist regulatory agencies in their screening efforts; however, to date, there has been little independent evaluation of these tools for this use. To contribute to such an evaluation, the International Life Sciences Institute Risk Science Institute (ILSI RSI) convened a Working Group to examine methodology used to construct statistically based SAR systems for developmental toxicity. This document reports on the deliberations of the Working Group, which concluded that an improved process is needed for utilizing developmental toxicity data in the construction of statistically based SAR models. The process must be objective, reproducible, rational and transparent. Moreover, it must be informed by the expertise of developmental toxicologists and biologists and must be subject to peer review.

In silico prediction of drug toxicity

It is essential, in order to minimise expensive drug failures due to toxicity being found in late development or even in clinical trials, to determine potential toxicity problems as early as possible. In view of the large libraries of compounds now being handled by combinatorial chemistry and high-throughput screening, identification of putative toxicity is advisable even before synthesis. Thus the use of predictive toxicology is called for. A number of in silico approaches to toxicity prediction are discussed. Quantitative structure-activity relationships (QSARs), relating mostly to specific chemical classes, have long been used for this purpose, and exist for a wide range of toxicity endpoints. However, QSARs also exist for the prediction of toxicity of very diverse libraries, although often such QSARs are of the classification type; that is, they predict simply whether or not a compound is toxic, and do not give an indication of the level of toxicity. Examples are given of all of these. A number of expert systems are available for toxicity prediction, most of them covering a range of toxicity endpoints. Those discussed include TOPKAT, CASE, DEREK, HazardExpert, OncoLogic and COMPACT. Comparative tests of the ability of these systems to predict carcinogenicity show that improvement is still needed. The consensus approach is recommended, whereby the results from several prediction systems are pooled.

The second National Toxicology Program comparative exercise on the prediction of rodent carcinogenicity: definitive results

Chemical carcinogenicity has been the target of a large array of attempts to create alternative predictive models, ranging from short-term biological assays (e.g. mutagenicity tests) to theoretical models. Among the theoretical models, the application of the science of structure-activity relationships (SAR) has earned special prominence. A crucial element is the independent evaluation of the predictive ability. In the past decade, there have been two fundamental comparative exercises on the prediction of chemical carcinogenicity, held under the aegis to the US National Toxicology Program (NTP). In both exercises, the predictions were published before the animal data were known, thus using a most stringent criterion of predictivity. We analyzed the results of the first comparative exercise in a previous paper [Mutat. Res. 387 (1997) 35]; here, we present the complete results of the second exercise, and we analyze and compare the prediction sets. The range of accuracy values was quite large: the systems that performed best in this prediction exercise were in the range 60-65% accuracy. They included various human experts approaches (e.g. Oncologic) and biologically based approaches (e.g. the experimental transformation assay in Syrian hamster embryo (SHE) cells). The main difficulty for the structure-activity relationship-based approaches was the discrimination between real carcinogens, and non-carcinogens containing structural alerts (SA) for genotoxic carcinogenicity. It is shown that the use of quantitative structure-activity relationship models, when possible, can contribute to overcome the above problem. Overall, given the uncertainty linked to the predictions, the predictions for the individual chemicals cannot be taken at face value; however, the general level of knowledge available today (especially for genotoxic carcinogens) allows qualified human experts to operate a very efficient priority setting of large sets of chemicals.

(Q)SARs: gatekeepers against risk on chemicals?

ECOSAR and DEREKfW predictions for the (eco)toxicological effects of circa 70 substances were compared with experimental data for risk assessment purposes. These and other (quantitative) structure-activity relationships ((Q)SARs) programs will play an important role in future chemical policies, such as in the European Union and The Netherlands, to reduce animal testing and costs and to speed up the number of risk assessments for hazardous chemicals. The two programs, ECOSAR and DEREKfW, were selected because they are easy to use and transparent in their predictions. They predict to which chemical class a substance belongs and also predict some (eco)toxicological properties. ECOSAR categorised 87% of the chemicals correctly in chemical classes. With regard to predicting ecotoxicity, criteria were drawn up for the reliability of the QSARs provided by ECOSAR. Application of these criteria had the result that half of the regression lines from ECOSAR were considered unreliable beforehand. It turned out, however, that the "unreliable" regression lines predicted similar accurately as the "reliable" lines, although much less chemicals were available for validating the "unreliable" QSARs. The overall accurate prediction of toxicity by ECOSAR was 67%. DEREKfW categorised 90% of the chemicals correctly in chemical classes, while 10% of the structural fragments needed a more detailed description. The accuracy of prediction was around 60% for sensitisation, 75% for genotoxicity and carcinogenicity for a limited number of chemicals. Irritation and reproductive toxicity were predicted poorly. Finally, it should be stressed that regulators and industries need to agree on the acceptability criteria relating to false negative and false positive (Q)SAR predictions. This to prevent unnecessary animal testing when regulators do not sufficiently rely on (Q)SAR predictions or to prevent too much faith in (Q)SAR predictions which will then may cause an insufficient protection of man and the environment. Therefore, if the regulatory trend is that (Q)SARs have to be applied more and more systematically in the risk assessment process, their validity and the available tools have to be explored further.

ADMET in silico modelling: towards prediction paradise?

Following studies in the late 1990s that indicated that poor pharmacokinetics and toxicity were important causes of costly late-stage failures in drug development, it has become widely appreciated that these areas should be considered as early as possible in the drug discovery process. However, in recent years, combinatorial chemistry and high-throughput screening have significantly increased the number of compounds for which early data on absorption, distribution, metabolism, excretion (ADME) and toxicity (T) are needed, which has in turn driven the development of a variety of medium and high-throughput in vitro ADMET screens. Here, we describe how in silico approaches will further increase our ability to predict and model the most relevant pharmacokinetic, metabolic and toxicity endpoints, thereby accelerating the drug discovery process.

Structure-activity relationships for the mutagenicity and carcinogenicity of simple and alpha-beta unsaturated aldehydes

Aldehydes are important industrial compounds that are used for the synthesis of chemicals and pharmaceuticals and as solvents, food additives, and disinfectants. Because of their reactivity, aldehydes are able to interact with electron-rich biological macromolecules and adverse health effects have been reported, including general toxicity, allergenic reactions, mutagenicity, and carcinogenicity. The cost, time, and number of animals necessary to adequately screen these chemicals places serious limitations on the number of aldehydes whose health potential can be studied and points to the need of using alternative methods for assessing, at least in a preliminary way, the risks associated with the use of aldehydes. A method of choice is the study of quantitative structure-activity relationships (QSARs). In the present work, we present QSAR models for the mutagenicity and carcinogenicity of simple aldehydes and alpha-beta unsaturated aldehydes. The models point to the role of electrophilicity, bulkiness, and hydrophobicity in the genotoxic activity of the aldehydes and lend themselves to the prediction of the activity of other untested chemicals of the same class.

The current status and future applicability of quantitative structure-activity relationships (QSARs) in predicting toxicity

The current status of quantitative structure-activity relationships (QSARs) in predicting toxicity is assessed. Widespread use of these methods to predict toxicity from chemical structure is possible, both by industry to develop new compounds, and also by regulatory agencies. The current use of QSARs is restricted by the lack of suitable toxicity data available for modelling, the suitability of simplistic modelling approaches for the prediction of certain endpoints, and the poor definition and utilisation of the applicability domain of models. Suggestions to resolve these issues are made.

Computer systems for the prediction of toxicity: an update

In order to survive in the current economic climate, the pharmaceutical, agrochemical and personal product companies are required to produce large numbers of new, effective products whilst significantly reducing development time and costs. With the advent of combinatorial chemistry and high-throughput screening (HTS), the numbers of new candidate structures coming out of the discovery cycle has increased significantly. This has created a demand for faster screening of the toxicological properties of these candidates. Not surprisingly, computer methods for toxicity prediction offer an attractive solution to this problem because of their ability to screen large numbers of structures even before synthesis has occurred. In this paper the major, commercially available computer software systems for toxicity prediction are discussed together with their main strengths and limitations.

Distributed structure-searchable toxicity (DSSTox) public database network: a proposal

The ability to assess the potential genotoxicity, carcinogenicity, or other toxicity of pharmaceutical or industrial chemicals based on chemical structure information is a highly coveted and shared goal of varied academic, commercial, and government regulatory groups. These diverse interests often employ different approaches and have different criteria and use for toxicity assessments, but they share a need for unrestricted access to existing public toxicity data linked with chemical structure information. Currently, there exists no central repository of toxicity information, commercial or public, that adequately meets the data requirements for flexible analogue searching, Structure-Activity Relationship (SAR) model development, or building of chemical relational databases (CRD). The distributed structure-searchable toxicity (DSSTox) public database network is being proposed as a community-supported, web-based effort to address these shared needs of the SAR and toxicology communities. The DSSTox project has the following major elements: (1) to adopt and encourage the use of a common standard file format (structure data file (SDF)) for public toxicity databases that includes chemical structure, text and property information, and that can easily be imported into available CRD applications; (2) to implement a distributed source approach, managed by a DSSTox Central Website, that will enable decentralized, free public access to structure-toxicity data files, and that will effectively link knowledgeable toxicity data sources with potential users of these data from other disciplines (such as chemistry, modeling, and computer science); and (3) to engage public/commercial/academic/industry groups in contributing to and expanding this community-wide, public data sharing and distribution effort. The DSSTox project's overall aims are to effect the closer association of chemical structure information with existing toxicity data, and to promote and facilitate structure-based exploration of these data within a common chemistry-based framework that spans toxicological disciplines.