Benchmarking Toxic Molecule Classification using Graph Neural Networks and Few Shot Learning

TL;DR

This paper presents a novel approach combining Graph Isomorphic Networks, Multi Headed Attention, and Few-Shot Learning to improve toxic molecule classification, achieving state-of-the-art results with limited data.

Contribution

It introduces a new methodology integrating advanced graph neural network techniques with Few-Shot Learning for better toxicity prediction.

Findings

Achieved an AUC-ROC of 0.816, surpassing baseline by 11.4%.

Demonstrated effectiveness of Few-Shot Learning in toxicology classification.

Validated approach on diverse toxicology datasets.

Abstract

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.