Identification of Robust Antibiotic Subgroups by Integrating Multi-Species Drug-Drug Interactions

Previous studies have shown that antibiotics can be dividedintogroups, and drug-drug interactions (DDI) depend on their groups.However, these studies focused on a specific bacteria strain (i.e., Escherichia coli BW25113). Existing datasets oftencontain noise. Noisy labeled data may have a bad effect on the clusteringresults. To address this problem, we developed a multi-source informationfusion method for integrating DDI information from multiple bacterialstrains. Specifically, we calculated drug similarities based on theDDI network of each bacterial strain and then fused these drug similaritymatrices to obtain a new fused similarity matrix. The fused similaritymatrix was combined with the T-distributed stochastic neighbor embeddingalgorithm, and hierarchical clustering algorithm can effectively identifyantibiotic subgroups. These antibiotic subgroups are strongly correlatedwith known antibiotic classifications, and group-group interactionsare almost monochromatic. In summary, our method provides a promisingframework for understanding the mechanism of action of antibioticsand exploring multi-species group-group interactions.