Prediction of antibacterial interaction between essential oils via graph embedding approach

Essential oils contain a variety of volatile metabolites, and are expected to be utilized in wide fields such as antimicrobials, insect repellents and herbicides. However, it is difficult to foresee the effect of mixing the oils because hundreds of compounds can be involved in synergistic and antagonistic interactions. For efficient formula optimization, we have developed and evaluated a machine learning method to classify antibacterial interactions between the oils. Cross-validation showed that graph embedding improved areas under the ROC curves for synergistic-versus-rest classification. Furthermore, antibacterial assay against Staphylococcus aureus revealed that oregano–ajowan, lemongrass–hiba, cinnamon–lemongrass and ajowan–ginger combinations exhibited synergistic interaction as predicted. These results indicate that graph embedding approach is useful for predicting synergistic interaction between antibacterial essential oils.