Genome-wide Identification of Type VI Secretion System Effectors via Machine Learning Uncovers Three New Bacterial Toxins

The type VI secretion system (T6SS) is a molecular harpoon used by 19% of sequenced bacteria to inject toxic proteins into microbial competitors or host cells. Despite its widespread distribution, many T6SS effectors (T6Es) remain unidentified in bacterial genomes. Here, we developed an XGBoost classifier integrating 42 genetic and biochemical features to identify novel T6Es. Applying our model to 2,559 T6SS-encoding bacterial genomes revealed 19,958 T6E candidates. We validated our computational predictions and confirmed that three effector candidates, having new 3D structures, demonstrated cytotoxicity to Escherichia coli and/or Saccharomyces cerevisiae , mostly in the bacterial periplasm. For two of these, cognate immunity proteins neutralizing cytotoxicity have been validated as well. T6EC11, a new peptidoglycan degrading effector, quickly leads to spherical E. coli , with inner membrane separation, and gradual peptidoglycan degradation in a unipolar manner, demonstrating a moon-like structural phenotype. T6EC12, a predicted cargo effector that is toxic to bacteria and yeast, carries a conserved zinc-binding motif. Mutations in the predicted catalytic center markedly reduced toxicity, suggesting that T6EC12 represents a novel metalloprotease T6E. Our study expands the known T6E repertoire and provides thousands of predictions of potential effectors in most T6SS-encoding bacteria.