Graph Clustering Based on Chemical Similarity in Marine Compounds and Antibacterial Compounds

Abstract

Many potential medicines have been discovered from marine natural products (MNPs). It indicates that marine compounds are essential sources in drug development and discovery. Although many marine compounds show particular biological activity, few are recorded as antibacterial compounds. Therefore, finding the potential compound as an antibacterial compound from a marine organism is still challenging. The aim of this study is to utilize a computational approach to discover potential antibacterial compounds from marine resources. The study focuses on employing the BiClusO algorithm for clustering based on the chemical similarity between marine compounds and antibacterial compounds. The results show that the number of clusters formed for marine biota compounds with antibiotic drug compounds is 4. Then the compounds of marine biota with antibiotic compounds formed 7 clusters. Finally, from these clusters, we obtain compounds that are predicted to have similar properties to antibacterial drugs or compounds. From 73 marine compounds, only Sarasinoside J and (-)-Sarasinoside K are predicted as potent antibacterial compounds.