Inferring human microbe–drug associations via multiple kernel fusion on graph neural network

作者：

Highlights：

• Our model applies the Graph Convolutional Network into Multiple Kernel fusion.

• We apply multi-layer GCN to extract different structural information in the graph.

• Our method has excellent performance on three microbe–drug association datasets.

• We discover some potential drugs being associated with COVID-19.

摘要

•Our model applies the Graph Convolutional Network into Multiple Kernel fusion.•We apply multi-layer GCN to extract different structural information in the graph.•Our method has excellent performance on three microbe–drug association datasets.•We discover some potential drugs being associated with COVID-19.

论文关键词：00-01,99-00,Microbe–drug association,Graph convolutional network,Multiple kernel fusion,Dual graph regularized least square,Bipartite network

论文评审过程：Received 21 September 2021, Revised 12 November 2021, Accepted 2 December 2021, Available online 10 December 2021, Version of Record 23 December 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107888