Prediction of Synergistic Antibiotic Combinations by Graph Learning

• DOI: 10.3389/fphar.2022.849006
• Publication Date: 2022-03-08T10:08:12Z
• Authors: Ji Lv; Guixia Liu; Yuan Ju; Ying Sun; Weiying Guo
• Keywords: Interpretability

Antibiotic resistance is a major public health concern. combinations, offering better efficacy at lower doses, are useful way to handle this problem. However, it difficult for us find effective antibiotic combinations in the vast chemical space. Herein, we propose graph learning framework predict synergistic combinations. In model, network proximity method combined with propagation was used quantify relationships of drug pairs, and found that tend have smaller proximity. Therefore, can be building an affinity matrix. Subsequently, matrix fed into regularization model potential Compared existing methods, our shows performance prediction interpretability.