Benchmarking Toxic Molecule Classification using Graph Neural Networks and Few Shot Learning
This work addresses toxic molecule classification for drug discovery and environmental risk assessment, representing an incremental improvement over existing methods.
The paper tackled the problem of toxic molecule classification with limited data and class imbalance by combining Graph Isomorphic Networks, Multi-Headed Attention, and adversarial augmentation with Few-Shot Learning, achieving a state-of-the-art AUC-ROC of 0.816, an 11.4% improvement over baseline GCNs.
Traditional methods like Graph Convolutional Networks (GCNs) face challenges with limited data and class imbalance, leading to suboptimal performance in graph classification tasks during toxicity prediction of molecules as a whole. To address these issues, we harness the power of Graph Isomorphic Networks, Multi Headed Attention and Free Large-scale Adversarial Augmentation separately on Graphs for precisely capturing the structural data of molecules and their toxicological properties. Additionally, we incorporate Few-Shot Learning to improve the model's generalization with limited annotated samples. Extensive experiments on a diverse toxicology dataset demonstrate that our method achieves an impressive state-of-art AUC-ROC value of 0.816, surpassing the baseline GCN model by 11.4%. This highlights the significance of our proposed methodology and Few Shot Learning in advancing Toxic Molecular Classification, with the potential to enhance drug discovery and environmental risk assessment processes.