Prediction of carcinogenicity for diverse chemicals based on substructure grouping and SVM modeling

Unlocking the potential of AI: Machine learning and deep learning models for predicting carcinogenicity of chemicals

The escalating apprehension surrounding the carcinogenic potential of chemicals emphasizes the imperative need for efficient methods of assessing carcinogenicity. Conventional experimental approaches such as in vitro and in vivo assays, albeit effective, suffer from being costly and time-consuming. In response to this challenge, new alternative methodologies, notably machine learning and deep learning techniques, have attracted attention for their potential in developing carcinogenicity prediction models. This article reviews the progress in predicting carcinogenicity using various machine learning and deep learning algorithms. A comparative analysis on these developed models reveals that support vector machine, random forest, and ensemble learning are commonly preferred for their robustness and effectiveness in predicting chemical carcinogenicity. Conversely, models based on deep learning algorithms, such as feedforward neural network, convolutional neural network, graph convolutional neural network, capsule neural network, and hybrid neural networks, exhibit promising capabilities but are limited by the size of available carcinogenicity datasets. This review provides a comprehensive analysis of current machine learning and deep learning models for carcinogenicity prediction, underscoring the importance of high-quality and large datasets. These observations are anticipated to catalyze future advancements in developing effective and generalizable machine learning and deep learning models for predicting chemical carcinogenicity.

Recent progress in machine learning approaches for predicting carcinogenicity in drug development

INTRODUCTION

This review explores the transformative impact of machine learning (ML) on carcinogenicity prediction within drug development. It discusses the historical context and recent advancements, emphasizing the significance of ML methodologies in overcoming challenges related to data interpretation, ethical considerations, and regulatory acceptance.

AREAS COVERED

The review comprehensively examines the integration of ML, deep learning, and diverse artificial intelligence (AI) approaches in various aspects of drug development safety assessments. It explores applications ranging from early-phase compound screening to clinical trial optimization, highlighting the versatility of ML in enhancing predictive accuracy and efficiency.

EXPERT OPINION

Through the analysis of traditional approaches such as in vivo rodent bioassays and in vitro assays, the review underscores the limitations and resource intensity associated with these methods. It provides expert insights into how ML offers innovative solutions to address these challenges, revolutionizing safety assessments in drug development.

DCAMCP: A deep learning model based on capsule network and attention mechanism for molecular carcinogenicity prediction

The carcinogenicity of drugs can have a serious impact on human health, so carcinogenicity testing of new compounds is very necessary before being put on the market. Currently, many methods have been used to predict the carcinogenicity of compounds. However, most methods have limited predictive power and there is still much room for improvement. In this study, we construct a deep learning model based on capsule network and attention mechanism named DCAMCP to discriminate between carcinogenic and non-carcinogenic compounds. We train the DCAMCP on a dataset containing 1564 different compounds through their molecular fingerprints and molecular graph features. The trained model is validated by fivefold cross-validation and external validation. DCAMCP achieves an average accuracy (ACC) of 0.718 ± 0.009, sensitivity (SE) of 0.721 ± 0.006, specificity (SP) of 0.715 ± 0.014 and area under the receiver-operating characteristic curve (AUC) of 0.793 ± 0.012. Meanwhile, comparable results can be achieved on an external validation dataset containing 100 compounds, with an ACC of 0.750, SE of 0.778, SP of 0.727 and AUC of 0.811, which demonstrate the reliability of DCAMCP. The results indicate that our model has made progress in cancer risk assessment and could be used as an efficient tool in drug design.

Predicting Dose-Range Chemical Toxicity using Novel Hybrid Deep Machine-Learning Method

Humans are exposed to thousands of chemicals, including environmental chemicals. Unfortunately, little is known about their potential toxicity, as determining the toxicity remains challenging due to the substantial resources required to assess a chemical in vivo. Here, we present a novel hybrid neural network (HNN) deep learning method, called HNN-Tox, to predict chemical toxicity at different doses. To develop a hybrid HNN-Tox method, we combined two neural network frameworks, the Convolutional Neural Network (CNN) and the multilayer perceptron (MLP)-type feed-forward neural network (FFNN). Combining the CNN and FCNN in the field of environmental chemical toxicity prediction is a novel approach. We developed several binary and multiclass classification models to assess dose-range chemical toxicity that is trained based on thousands of chemicals with known toxicity. The performance of the HNN-Tox was compared with other machine-learning methods, including Random Forest (RF), Bootstrap Aggregation (Bagging), and Adaptive Boosting (AdaBoost). We also analyzed the model performance dependency on varying features, descriptors, dataset size, route of exposure, and toxic dose. The HNN-Tox model, trained on 59,373 chemicals annotated with known LD50 and routes of exposure, maintained its predictive ability with an accuracy of 84.9% and 84.1%, even after reducing the descriptor size from 318 to 51, and the area under the ROC curve (AUC) was 0.89 and 0.88, respectively. Further, we validated the HNN-Tox with several external toxic chemical datasets on a large scale. The HNN-Tox performed optimally or better than the other machine-learning methods for diverse chemicals. This study is the first to report a large-scale prediction of dose-range chemical toxicity with varying features. The HNN-Tox has broad applicability in predicting toxicity for diverse chemicals and could serve as an alternative methodology approach to animal-based toxicity assessment.

Predicting Chemical Carcinogens Using a Hybrid Neural Network Deep Learning Method

Determining environmental chemical carcinogenicity is urgently needed as humans are increasingly exposed to these chemicals. In this study, we developed a hybrid neural network (HNN) method called HNN-Cancer to predict potential carcinogens of real-life chemicals. The HNN-Cancer included a new SMILES feature representation method by modifying our previous 3D array representation of 1D SMILES simulated by the convolutional neural network (CNN). We developed binary classification, multiclass classification, and regression models based on diverse non-congeneric chemicals. Along with the HNN-Cancer model, we developed models based on the random forest (RF), bootstrap aggregating (Bagging), and adaptive boosting (AdaBoost) methods for binary and multiclass classification. We developed regression models using HNN-Cancer, RF, support vector regressor (SVR), gradient boosting (GB), kernel ridge (KR), decision tree with AdaBoost (DT), KNeighbors (KN), and a consensus method. The performance of the models for all classifications was assessed using various statistical metrics. The accuracy of the HNN-Cancer, RF, and Bagging models were 74%, and their AUC was ~0.81 for binary classification models developed with 7994 chemicals. The sensitivity was 79.5% and the specificity was 67.3% for the HNN-Cancer, which outperforms the other methods. In the case of multiclass classification models with 1618 chemicals, we obtained the optimal accuracy of 70% with an AUC 0.7 for HNN-Cancer, RF, Bagging, and AdaBoost, respectively. In the case of regression models, the correlation coefficient (R) was around 0.62 for HNN-Cancer and RF higher than the SVM, GB, KR, DTBoost, and NN machine learning methods. Overall, the HNN-Cancer performed better for the majority of the known carcinogen experimental datasets. Further, the predictive performance of HNN-Cancer on diverse chemicals is comparable to the literature-reported models that included similar and less diverse molecules. Our HNN-Cancer could be used in identifying potentially carcinogenic chemicals for a wide variety of chemical classes.

DeepCarc: Deep Learning-Powered Carcinogenicity Prediction Using Model-Level Representation

Carcinogenicity testing plays an essential role in identifying carcinogens in environmental chemistry and drug development. However, it is a time-consuming and label-intensive process to evaluate the carcinogenic potency with conventional 2-years rodent animal studies. Thus, there is an urgent need for alternative approaches to providing reliable and robust assessments on carcinogenicity. In this study, we proposed a DeepCarc model to predict carcinogenicity for small molecules using deep learning-based model-level representations. The DeepCarc Model was developed using a data set of 692 compounds and evaluated on a test set containing 171 compounds in the National Center for Toxicological Research liver cancer database (NCTRlcdb). As a result, the proposed DeepCarc model yielded a Matthews correlation coefficient (MCC) of 0.432 for the test set, outperforming four advanced deep learning (DL) powered quantitative structure-activity relationship (QSAR) models with an average improvement rate of 37%. Furthermore, the DeepCarc model was also employed to screen the carcinogenicity potential of the compounds from both DrugBank and Tox21. Altogether, the proposed DeepCarc model could serve as an early detection tool (https://github.com/TingLi2016/DeepCarc) for carcinogenicity assessment.

CarcinoPred-EL: Novel models for predicting the carcinogenicity of chemicals using molecular fingerprints and ensemble learning methods

Carcinogenicity refers to a highly toxic end point of certain chemicals, and has become an important issue in the drug development process. In this study, three novel ensemble classification models, namely Ensemble SVM, Ensemble RF, and Ensemble XGBoost, were developed to predict carcinogenicity of chemicals using seven types of molecular fingerprints and three machine learning methods based on a dataset containing 1003 diverse compounds with rat carcinogenicity. Among these three models, Ensemble XGBoost is found to be the best, giving an average accuracy of 70.1 ± 2.9%, sensitivity of 67.0 ± 5.0%, and specificity of 73.1 ± 4.4% in five-fold cross-validation and an accuracy of 70.0%, sensitivity of 65.2%, and specificity of 76.5% in external validation. In comparison with some recent methods, the ensemble models outperform some machine learning-based approaches and yield equal accuracy and higher specificity but lower sensitivity than rule-based expert systems. It is also found that the ensemble models could be further improved if more data were available. As an application, the ensemble models are employed to discover potential carcinogens in the DrugBank database. The results indicate that the proposed models are helpful in predicting the carcinogenicity of chemicals. A web server called CarcinoPred-EL has been built for these models (http://ccsipb.lnu.edu.cn/toxicity/CarcinoPred-EL/).

In Silico Estimation of Chemical Carcinogenicity with Binary and Ternary Classification Methods

Carcinogenicity is one of the most concerned properties of chemicals to human health, thus it is important to identify chemical carcinogenicity as early as possible. In this study, 829 diverse compounds with rat carcinogenicity were collected from Carcinogenic Potency Database (CPDB). Using six types of fingerprints to represent the molecules, 30 binary and ternary classification models were generated to predict chemical carcinogenicity by five machine learning methods. The models were evaluated by an external validation set containing 87 chemicals from ISSCAN database. The best binary model was developed by MACCS keys and kNN algorithm with predictive accuracy at 83.91 %, while the best ternary model was also generated by MACCS keys and kNN algorithm with overall accuracy at 80.46 %. Furthermore, the best binary and ternary classification models were used to estimate carcinogenicity of tobacco smoke components containing 2251 compounds. 981 ones were predicted as carcinogens by binary classification model, while 110 compounds were predicted as strong carcinogens and 807 ones as weak carcinogens by ternary classification model. The results indicated that our models would be helpful for prediction of chemical carcinogenicity.

Support vector machine (SVM) classification model based rational design of novel tetronic acid derivatives as potent insecticidal and acaricidal agents

A novel SVM classification model was constructed and applied in the development of novel tetronic acid derivatives as potent insecticidal and acaricidal agents.

A clinical risk stratification tool for predicting treatment resistance in major depressive disorder

BACKGROUND

Early identification of depressed individuals at high risk for treatment resistance could be helpful in selecting optimal setting and intensity of care. At present, validated tools to facilitate this risk stratification are rarely used in psychiatric practice.

METHODS

Data were drawn from the first two treatment levels of a multicenter antidepressant effectiveness study in major depressive disorder, the STAR*D (Sequenced Treatment Alternatives to Relieve Depression) cohort. This cohort was divided into training, testing, and validation subsets. Only clinical or sociodemographic variables available by or readily amenable to self-report were considered. Multivariate models were developed to discriminate individuals reaching remission with a first or second pharmacological treatment trial from those not reaching remission despite two trials.

RESULTS

A logistic regression model achieved an area under the receiver operating characteristic curve exceeding .71 in training, testing, and validation cohorts and maintained good calibration across cohorts. Performance of three alternative models with machine learning approaches--a naïve Bayes classifier and a support vector machine, and a random forest model--was less consistent. Similar performance was observed between more and less severe depression, men and women, and primary versus specialty care sites. A web-based calculator was developed that implements this tool and provides graphical estimates of risk.

CONCLUSION

Risk for treatment resistance among outpatients with major depressive disorder can be estimated with a simple model incorporating baseline sociodemographic and clinical features. Future studies should examine the performance of this model in other clinical populations and its utility in treatment selection or clinical trial design.

Improvement of carcinogenicity prediction performances based on sensitivity analysis in variable selection of SVM models

A new sensitivity analysis (SA) method for variable selection in support vector machine (SVM) was proposed to improve the performance level of the QSAR model to predict carcinogenicity based on the correlation coefficient (CC) method used in our preceding study. The performances of both methods were also compared with that of the F-score (FS) method proposed by Chang and Lin. The 911 non-congeneric chemicals were classified into 20 mutually overlapping groups according to contained substructures, and a specific SVM model created on chemicals belonging to each group was optimized by searching the best set of SVM parameters while successively omitting descriptors of lower absolute values of sensitivity, CC or FS until the maximum predictive performance was obtained. The SA method improves the overall accuracy from 80% of CC and FS to 84%, which is considerably higher than those of existing models for predicting the carcinogenicity of non-congeneric chemicals. It selects the optimum sets of effective descriptors fewer than the CC and FS methods, and is not time-consuming and can be applied to a large set of initial descriptors. It is concluded that SA is superior as a variable selection method in SVM models.

Classification models for safe drug molecules

Frequent failure of drug candidates during development stages remains the major deterrent for an early introduction of new drug molecules. The drug toxicity is the major cause of expensive late-stage development failures. An early identification/optimization of the most favorable molecule will naturally save considerable cost, time, human efforts and minimize animal sacrifice. (Quantitative) Structure Activity Relationships [(Q)SARs] represent statistically derived predictive models correlating biological activity (including desirable therapeutic effect and undesirable side effects) of chemicals (drugs/toxicants/environmental pollutants) with molecular descriptors and/or properties. (Q)SAR models which categorize the available data into two or more groups/classes are known as classification models. Numerous techniques of diverse nature are being presently employed for development of classification models. Though there is an increasing use of classification models for prediction of either biological activity or toxicity, the future trend will naturally be towards the development of classification models capable of simultaneous prediction of biological activity, toxicity, and pharmacokinetic parameters so as to accelerate development of bioavailable safe drug molecules.

Which is a more accurate predictor in colorectal survival analysis? Nine data mining algorithms vs. the TNM staging system

OBJECTIVE

Over the past decades, many studies have used data mining technology to predict the 5-year survival rate of colorectal cancer, but there have been few reports that compared multiple data mining algorithms to the TNM classification of malignant tumors (TNM) staging system using a dataset in which the training and testing data were from different sources. Here we compared nine data mining algorithms to the TNM staging system for colorectal survival analysis.

METHODS

Two different datasets were used: 1) the National Cancer Institute's Surveillance, Epidemiology, and End Results dataset; and 2) the dataset from a single Chinese institution. An optimization and prediction system based on nine data mining algorithms as well as two variable selection methods was implemented. The TNM staging system was based on the 7(th) edition of the American Joint Committee on Cancer TNM staging system.

RESULTS

When the training and testing data were from the same sources, all algorithms had slight advantages over the TNM staging system in predictive accuracy. When the data were from different sources, only four algorithms (logistic regression, general regression neural network, bayesian networks, and Naïve Bayes) had slight advantages over the TNM staging system. Also, there was no significant differences among all the algorithms (p>0.05).

CONCLUSIONS

The TNM staging system is simple and practical at present, and data mining methods are not accurate enough to replace the TNM staging system for colorectal cancer survival prediction. Furthermore, there were no significant differences in the predictive accuracy of all the algorithms when the data were from different sources. Building a larger dataset that includes more variables may be important for furthering predictive accuracy.

High-Dimensional descriptor selection and computational QSAR modeling for antitumor activity of ARC-111 analogues Based on Support Vector Regression (SVR)

To design ARC-111 analogues with improved efficiency, we constructed the QSAR of 22 ARC-111 analogues with RPMI8402 tumor cells. First, the optimized support vector regression (SVR) model based on the literature descriptors and the worst descriptor elimination multi-roundly (WDEM) method had similar generalization as the artificial neural network (ANN) model for the test set. Secondly, seven and 11 more effective descriptors out of 2,923 features were selected by the high-dimensional descriptor selection nonlinearly (HDSN) and WDEM method, and the SVR models (SVR3 and SVR4) with these selected descriptors resulted in better evaluation measures and a more precise predictive power for the test set. The interpretability system of better SVR models was further established. Our analysis offers some useful parameters for designing ARC-111 analogues with enhanced antitumor activity.

On the information expressed in enzyme structure: more lessons from ribonuclease A

Brownian computations were directed at Ribonuclease A (RNase A) and variants in folded states so as to quantify information of the statistical type at the atom/covalent bond level. This advanced the research reported in this journal last year on the information properties of enzyme primary structure. Brownian computation data are illustrated for a sixteen-member library. The results identify signature traits that distinguish the folded wild type (WT) molecule from variants. The distinctions are explainable in terms of correlated information and dispersion energy. The Brownian tools used for this study can be directed at other protein families (e.g., kinases, isomerases, etc.) in rapid screening, QSAR, and design applications.