A few-shot link prediction framework to drug repurposing using multi-level attention network

Chenglin Yang, Xianlai Chen, Jincai Huang, Ying An, Zhenyu Huang, Yu Sun

Abstract: Highlights•We propose a novel meta-learning framework for drug repurposing based on MLAN.•The entity-level attention of MLAN captures valuable neighborhood information.•The learner of the triplet-level attention conducts meta-relation learning.•MTransD optimizes each meta-task within a unified framework.•We validate the potential of our method in drug repurposing for Covid-19.