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.

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.

This study addresses the challenge of assessing the carcinogenic potential of hazardous chemical mixtures, such as per- and polyfluorinated substances (PFASs), which are known to contribute significantly to cancer development. Here, we propose a novel framework called HNNMixCancer that utilizes a hybrid neural network (HNN) integrated into a machine-learning framework. This framework incorporates a mathematical model to simulate chemical mixtures, enabling the creation of classification models for binary (carcinogenic or noncarcinogenic) and multiclass classification (categorical carcinogenicity) and regression (carcinogenic potency). Through extensive experimentation, we demonstrate that our HNN model outperforms other methodologies, including random forest, bootstrap aggregating, adaptive boosting, support vector regressor, gradient boosting, kernel ridge, decision tree with AdaBoost, and KNeighbors, achieving a superior accuracy of 92.7% in binary classification. To address the limited availability of experimental data and enrich the training data, we generate an assumption-based virtual library of chemical mixtures using a known carcinogenic and noncarcinogenic single chemical for all the classification models. Remarkably, in this case, all methods achieve accuracies exceeding 98% for binary classification. In external validation tests, our HNN method achieves the highest accuracy of 80.5%. Furthermore, in multiclass classification, the HNN demonstrates an overall accuracy of 96.3%, outperforming RF, Bagging, and AdaBoost, which achieved 91.4%, 91.7%, and 80.2%, respectively. In regression models, HNN, RF, SVR, GB, KR, DT with AdaBoost, and KN achieved average R2 values of 0.96, 0.90, 0.77, 0.94, 0.96, 0.96, and 0.97, respectively, showcasing their effectiveness in predicting the concentration at which a chemical mixture becomes carcinogenic. Our method exhibits exceptional predictive power in prioritizing carcinogenic chemical mixtures, even when relying on assumption-based mixtures. This capability is particularly valuable for toxicology studies that lack experimental data on the carcinogenicity and toxicity of chemical mixtures. To our knowledge, this study introduces the first method for predicting the carcinogenic potential of chemical mixtures. The HNNMixCancer framework offers a novel alternative for dose-dependent carcinogen prediction. Ongoing efforts involve implementing the HNN method to predict mixture toxicity and expanding the application of HNNMixCancer to include multiple mixtures such as PFAS mixtures and co-occurring chemicals.

Humans are exposed to thousands of potentially toxic chemicals including environmental chemicals such as industrial wastes, food products, solvents, air pollutants, fertilizers, pesticides, insecticides, carcinogens, drugs, metals/metalloids, and other industrial chemicals. Approximately 300,000 such chemicals currently in use, unfortunately little is known about their potential toxicity. Determining human toxicity potential of chemicals remains a challenge due to a substantial resource required to assess a chemical in-vivo, and only a few thousand single chemicals in commercial use has been evaluated. In this study, to predict the environmental chemical toxicity, we developed a new hybrid neural network (HNN) deep learning model consisting of a Convolutional Neural Network (CNN) and multilayer perceptron (MLP) type feed forward neural network (FFNN). Our HNN deep learning model trained based on thousands of chemicals, presented the best performance for majority of the cases. Taken together, our hybrid HNN deep learning models has a wide applicability in the prediction of toxicity of any chemical category and its mixtures.

Accurately predicting ligand binding affinity in a virtual screening campaign is still challenging. Here, we developed hybrid neural network (HNN) machine deep learning methods, HNN-denovo and HNN-affinity, by combining the 3D-CNN (convolutional neural network) and the FFNN (fast forward neural network) hybrid neural network framework. The HNN-denovo uses protein pocket structure and protein–ligand interactions as input features. The HNN-affinity uses protein sequences and ligand features as input features. The HNN method combines the CNN and FCNN machine architecture for the protein structure or protein sequence and ligand descriptors. To train the model, the HNN methods used thousands of known protein–ligand binding affinity data retrieved from the PDBBind database. We also developed the Random Forest (RF), Gradient Boosting (GB), Decision Tree with AdaBoost (DT), and a consensus model. We compared the HNN results with models developed based on the RF, GB, and DT methods. We also independently compared the HNN method results with the literature reported deep learning protein–ligand binding affinity predictions made by the DLSCORE, KDEEP, and DeepAtom. The predictive performance of the HNN methods (max Pearson’s R achieved was 0.86) was consistently better than or comparable to the DLSCORE, KDEEP, and DeepAtom deep learning learning methods for both balanced and unbalanced data sets. The HNN-affinity can be applied for the protein–ligand affinity prediction even in the absence of protein structure information, as it considers the protein sequence as standalone feature in addition to the ligand descriptors. The HNN-denovo method can be efficiently implemented to the structure-based de novo drug design campaign. The HNN-affinity method can be used in conjunction with the deep learning molecular docking protocols as a standalone. Further, it can be combined with the conventional molecular docking methods as a multistep approach to rapidly screen billions of diverse compounds. The HNN method are highly scalable in the cloud ML platform.