Predicting Environmental Chemical Carcinogenicity using a Hybrid Machine-Learning Approach

Limbu, Sarita; Dakshanamurthy, Sivanesan

doi:10.1101/2021.05.03.442477

Cited by 3 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Instead, we utilize a dataset used by CarcinoPred-EL to design a comparable method and evaluate it on the continuous CPDB data. Furthermore we design an AdaBoost decision tree model emulating methodology from the work of Limbu and Dakshanamurthy (2021) . We also compare on an external CCRIS test set where our model outperforms all variants of CarcinoPred-EL the AdaBoost decision tree ( Table 2 ).…”

Section: Resultsmentioning

confidence: 99%

“…A number of approaches use cheminformatics features [such as molecular descriptors ( Landrum, 2016 ; Moriwaki et al , 2018 )] to construct regression models for continuous carcinogenicity prediction ( Fjodorova et al , 2010 ; Singh et al , 2013 ; Wu et al , 2015 ). Recently, Limbu and Dakshanamurthy (2021) compared various regression models of these sorts, of which AdaBoost ( Freund and Schapire, 1997 ) was found to be among the most accurate for dose-rate prediction.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A graph neural network approach for molecule carcinogenicity prediction

et al. 2022

View full text Add to dashboard Cite

Motivation Molecular carcinogenicity is a preventable cause of cancer, but systematically identifying carcinogenic compounds, which involves performing experiments on animal models, is expensive, time consuming and low throughput. As a result, carcinogenicity information is limited and building data-driven models with good prediction accuracy remains a major challenge. Results In this work, we propose CONCERTO, a deep learning model that uses a graph transformer in conjunction with a molecular fingerprint representation for carcinogenicity prediction from molecular structure. Special efforts have been made to overcome the data size constraint, such as multi-round pre-training on related but lower quality mutagenicity data, and transfer learning from a large self-supervised model. Extensive experiments demonstrate that our model performs well and can generalize to external validation sets. CONCERTO could be useful for guiding future carcinogenicity experiments and provide insight into the molecular basis of carcinogenicity. Availability and implementation The code and data underlying this article are available on github at https://github.com/bowang-lab/CONCERTO

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

A graph neural network approach for molecule carcinogenicity prediction

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Then we concatenate the CNN and FFNN models but not their individual predictions to make the final prediction. We included a new SMILES feature representation method in the HNN framework by modifying our previous 3D array representation of 1D SMILES simulated by the convolutional neural network (CNN) [32,33]. We assembled protein-ligand binding data such as Kd, Ki, and the combined Kd and Ki values from a general set and refined set obtained from the PDBBind database.…”

Section: Resultsmentioning

confidence: 99%

A New Hybrid Neural Network Deep Learning Method for Protein–Ligand Binding Affinity Prediction and De Novo Drug Design

Limbu

Dakshanamurthy

2022

IJMS

Self Cite

View full text Add to dashboard Cite

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.

show abstract

“…15,29,30,130−134 Limbu and Dakshanamurthy built carcinogenicity prediction models using a hybrid NN architecture with three data sets containing more than 10,000 chemicals, and 653 molecular descriptors. 130 The hybrid NN achieved an average ACC of 74.3% and mean ROC-AUC of 80.6%, which were superior to those of the AdaBoost model but not those of the Bagging and RF models using the same data. One of the leading causes of low predictive accuracy of DL models is sparse data sets.…”

Section: Carcinogenesis Predictionmentioning

confidence: 85%

“…In recent years, many AI-based models and tools have been developed to predict the carcinogenic potential of compounds. ,,,− Limbu and Dakshanamurthy built carcinogenicity prediction models using a hybrid NN architecture with three data sets containing more than 10,000 chemicals, and 653 molecular descriptors . The hybrid NN achieved an average ACC of 74.3% and mean ROC-AUC of 80.6%, which were superior to those of the AdaBoost model but not those of the Bagging and RF models using the same data.…”

Section: Recent Advances In Ai-based Drug Toxicity Predictionmentioning

confidence: 99%

Artificial Intelligence in Drug Toxicity Prediction: Recent Advances, Challenges, and Future Perspectives

Tran

Wibowo

Tayara

et al. 2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Toxicity prediction is a critical step in the drug discovery process that helps identify and prioritize compounds with the greatest potential for safe and effective use in humans, while also reducing the risk of costly late-stage failures. It is estimated that over 30% of drug candidates are discarded owing to toxicity. Recently, artificial intelligence (AI) has been used to improve drug toxicity prediction as it provides more accurate and efficient methods for identifying the potentially toxic effects of new compounds before they are tested in human clinical trials, thus saving time and money. In this review, we present an overview of recent advances in AI-based drug toxicity prediction, including the use of various machine learning algorithms and deep learning architectures, of six major toxicity properties and Tox21 assay end points. Additionally, we provide a list of public data sources and useful toxicity prediction tools for the research community and highlight the challenges that must be addressed to enhance model performance. Finally, we discuss future perspectives for AI-based drug toxicity prediction. This review can aid researchers in understanding toxicity prediction and pave the way for new methods of drug discovery.

show abstract

Predicting Environmental Chemical Carcinogenicity using a Hybrid Machine-Learning Approach

Cited by 3 publications

References 21 publications

A graph neural network approach for molecule carcinogenicity prediction

A graph neural network approach for molecule carcinogenicity prediction

A New Hybrid Neural Network Deep Learning Method for Protein–Ligand Binding Affinity Prediction and De Novo Drug Design

Artificial Intelligence in Drug Toxicity Prediction: Recent Advances, Challenges, and Future Perspectives

Contact Info

Product

Resources

About