Related Experiment Video
Updated: Jun 27, 2025
Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers
Published on: March 1, 2024
GraphEGFR: Multi-task and transfer learning based on molecular graph attention mechanism and fingerprints improving
Bundit Boonyarit1, Nattawin Yamprasert2, Pawit Kaewnuratchadasorn3
1School of Information Science and Technology, Vidyasirimedhi Institute of Science and Technology, Rayong, Thailand.
Abstract:
The proteins within the human epidermal growth factor receptor (EGFR) family, members of the tyrosine kinase receptor family, play a pivotal role in the molecular mechanisms driving the development of various tumors. Tyrosine kinase inhibitors, key compounds in targeted therapy, encounter challenges in cancer treatment due to emerging drug resistance mutations. Consequently, machine learning has undergone significant evolution to address the challenges of cancer drug discovery related to EGFR family proteins. However, the application of deep learning in this area is hindered by inherent difficulties associated with small-scale data, particularly the risk of overfitting. Moreover, the design of a model architecture that facilitates learning through multi-task and transfer learning, coupled with appropriate molecular representation, poses substantial challenges. In this study, we introduce GraphEGFR, a deep learning regression model designed to enhance molecular representation and model architecture for predicting the bioactivity of inhibitors against both wild-type and mutant EGFR family proteins. GraphEGFR integrates a graph attention mechanism for molecular graphs with deep and convolutional neural networks for molecular fingerprints. We observed that GraphEGFR models employing multi-task and transfer learning strategies generally achieve predictive performance comparable to existing competitive methods. The integration of molecular graphs and fingerprints adeptly captures relationships between atoms and enables both global and local pattern recognition. We further validated potential multi-targeted inhibitors for wild-type and mutant HER1 kinases, exploring key amino acid residues through molecular dynamics simulations to understand molecular interactions. This predictive model offers a robust strategy that could significantly contribute to overcoming the challenges of developing deep learning models for drug discovery with limited data and exploring new frontiers in multi-targeted kinase drug discovery for EGFR family proteins.
Insights
GraphEGFR, a novel deep learning model, enhances prediction of drug activity against EGFR family proteins, overcoming data limitations for improved cancer drug discovery.
Area of Science:
- Computational chemistry and bioinformatics
- Machine learning in drug discovery
- Oncology and molecular biology
Background:
- Epidermal growth factor receptor (EGFR) family proteins are crucial in tumor development.
- Tyrosine kinase inhibitors face challenges due to drug resistance mutations.
- Deep learning for cancer drug discovery is limited by small datasets and overfitting.
Purpose of the Study:
- To introduce GraphEGFR, a deep learning model for predicting bioactivity of inhibitors against wild-type and mutant EGFR family proteins.
- To enhance molecular representation and model architecture for improved drug discovery predictions.
- To address challenges of limited data and overfitting in deep learning for kinase inhibitors.
Main Methods:
- Developed GraphEGFR, integrating graph attention mechanisms with deep and convolutional neural networks.
- Employed multi-task and transfer learning strategies within the GraphEGFR model.
- Utilized molecular graphs and fingerprints for comprehensive molecular representation.
- Performed molecular dynamics simulations to explore interactions of potential multi-targeted inhibitors.
Main Results:
- GraphEGFR models achieved predictive performance comparable to existing methods using multi-task and transfer learning.
- The integration of molecular graphs and fingerprints effectively captured atomic relationships and patterns.
- Identified potential multi-targeted inhibitors for wild-type and mutant HER1 kinases.
Conclusions:
- GraphEGFR offers a robust strategy for developing deep learning models in drug discovery with limited data.
- The model advances multi-targeted kinase drug discovery for the EGFR family.
- This approach aids in understanding molecular interactions for inhibitor design.
