关键词: 深度学习, 分子毒性预测, 元学习, 图同构网络, Transformer

Abstract: Molecular toxicity prediction was an important task in computer-aided Drug development. Neural networks can be used to predict molecular toxicity and safety. To obtain more accurate toxicity prediction results, a molecular toxicity prediction model based on meta graph isomorphism network was proposed, namely Meta-MTP. Firstly, graph isomorphism neural network was used to obtain molecular characterization by using atoms as nodes, bonds as edges, and molecules as graph structures. The pre-trained model was used to initialize the Graph Isomorphism Network (GIN) to obtain better parameters. A feedforward Transformer incorporated hierarchical attention and local enhancement was introduced. Atom type prediction and bond prediction were used as auxiliary tasks to extract more internal molecular information. The model was trained through a meta learning dual level optimization strategy. Finally, the model was trained using TOX21 and SIDER datasets. Experimental results show that Meta-MTP has good performance and molecular toxicity prediction ability. When the number of samples is 10, compared to FSGNNTR (Few-Shot Graph Neural Network-TRansformer) model in all tasks, the Area Under the Curve (AUC) is improved by 1.4% and 5.4%. Compared to GIN, Graph Convolutional Network (GCN), and GraphSAGE three traditional graph neural network models, the AUC improves by 18.3% to 23.7% and 7.3% to 22.2% respectively.

Key words: deep learning, molecular toxicity prediction, meta learning, graph isomorphism network, Transformer