MSFF-MA-DDI: Multi-Source Feature Fusion with Multiple Attention blocks for predicting Drug-Drug Interaction events

Qi Jin, Jiang Xie, Dingkai Huang, Chang Zhao, Hongjian He

Abstract: Highlights•A multi-source feature fusion method named MSFF-MA-DDI is proposed for predicting DDI events.•Long-distance atoms relationships reflect the global features of the drug molecule sequence.•A self-attention block based on position embedding is built to obtain drug sequence features.•The experimental results show that MSFF-MA-DDI achieves good performance.