Predicting Dose-Dependent Carcinogenicity of Chemical Mixtures Using a Novel Hybrid Neural Network Framework and Mathematical Approach
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 HNN<sub>MixCancer</sub...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Toxics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2305-6304/11/7/605 |
| _version_ | 1797587378803572736 |
|---|---|
| author | Sarita Limbu Sivanesan Dakshanamurthy |
| author_facet | Sarita Limbu Sivanesan Dakshanamurthy |
| author_sort | Sarita Limbu |
| collection | DOAJ |
| description | 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 HNN<sub>MixCancer</sub> 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 R<sup>2</sup> 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 HNN<sub>MixCancer</sub> 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 HNN<sub>MixCancer</sub> to include multiple mixtures such as PFAS mixtures and co-occurring chemicals. |
| first_indexed | 2024-03-11T00:36:10Z |
| format | Article |
| id | doaj.art-667881af9eb0410698eacd79d89eff02 |
| institution | Directory Open Access Journal |
| issn | 2305-6304 |
| language | English |
| last_indexed | 2024-03-11T00:36:10Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Toxics |
| spelling | doaj.art-667881af9eb0410698eacd79d89eff022023-11-18T21:37:14ZengMDPI AGToxics2305-63042023-07-0111760510.3390/toxics11070605Predicting Dose-Dependent Carcinogenicity of Chemical Mixtures Using a Novel Hybrid Neural Network Framework and Mathematical ApproachSarita Limbu0Sivanesan Dakshanamurthy1Lombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, DC 20057, USALombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, DC 20057, USAThis 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 HNN<sub>MixCancer</sub> 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 R<sup>2</sup> 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 HNN<sub>MixCancer</sub> 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 HNN<sub>MixCancer</sub> to include multiple mixtures such as PFAS mixtures and co-occurring chemicals.https://www.mdpi.com/2305-6304/11/7/605hybrid neural networkdose-dependent carcinogenicitychemical mixturesmachine learningper- and polyfluorinated substances |
| spellingShingle | Sarita Limbu Sivanesan Dakshanamurthy Predicting Dose-Dependent Carcinogenicity of Chemical Mixtures Using a Novel Hybrid Neural Network Framework and Mathematical Approach Toxics hybrid neural network dose-dependent carcinogenicity chemical mixtures machine learning per- and polyfluorinated substances |
| title | Predicting Dose-Dependent Carcinogenicity of Chemical Mixtures Using a Novel Hybrid Neural Network Framework and Mathematical Approach |
| title_full | Predicting Dose-Dependent Carcinogenicity of Chemical Mixtures Using a Novel Hybrid Neural Network Framework and Mathematical Approach |
| title_fullStr | Predicting Dose-Dependent Carcinogenicity of Chemical Mixtures Using a Novel Hybrid Neural Network Framework and Mathematical Approach |
| title_full_unstemmed | Predicting Dose-Dependent Carcinogenicity of Chemical Mixtures Using a Novel Hybrid Neural Network Framework and Mathematical Approach |
| title_short | Predicting Dose-Dependent Carcinogenicity of Chemical Mixtures Using a Novel Hybrid Neural Network Framework and Mathematical Approach |
| title_sort | predicting dose dependent carcinogenicity of chemical mixtures using a novel hybrid neural network framework and mathematical approach |
| topic | hybrid neural network dose-dependent carcinogenicity chemical mixtures machine learning per- and polyfluorinated substances |
| url | https://www.mdpi.com/2305-6304/11/7/605 |
| work_keys_str_mv | AT saritalimbu predictingdosedependentcarcinogenicityofchemicalmixturesusinganovelhybridneuralnetworkframeworkandmathematicalapproach AT sivanesandakshanamurthy predictingdosedependentcarcinogenicityofchemicalmixturesusinganovelhybridneuralnetworkframeworkandmathematicalapproach |