Table 2 Different methods to predict microbe–drug association
From: In-silico computational approaches to study microbiota impacts on diseases and pharmacotherapy
Category | Method | Description | Refs. |
|---|---|---|---|
Graph Convolutional Network (GCN) | GCNMDA | A conditional random context (CRF) and a CRF layer focus function in the hidden GCN layer are used to ensure that the same nodes have the same representations | [61] |
Ensembling graph attention | EGATMDA | To learn embedded nodes for microbes and drugs, a convolutional graph network is built at the node level for each input diagram | [55] |
Heterogeneous network embedding representation | HNERMDA | Metapath2vec has developed a heterogeneous network display learning approach for learning low-embedded microbe and drug displays in this context | [56] |
Multi-modal variational graph embedding | Graph2MDA | A graph with variations A deep neural network classifier was used to predict microbe–drug relationships after an auto encoder was trained to learn the informative and interpretable latent representations of each node and the whole graph | [67] |
Based on KATZ measurements | KATZ | By bringing the chemical structures together and the similarity of the nucleus, they have created the Gaussian interaction profile of the drug unification network | [57] |