PTSomics - Machine Learning-Driven Discovery and Analysis of Polymorphic Toxin and Effector Systems - PROJECT SUMMARY/ABSTRACT Bacterial protein toxins and effectors play a pivotal role in organismal conflicts, contributing significantly to self/non-self recognition, bacteria-phage interactions, and the pathogenesis of bacterial diseases in human and non-human animals, as well as commercially important plants impacting food security. Propelled by the constant evolution of an arms race, these proteins and their associated components typically evolve rapidly and display enormous diversity in terms of protein sequence, structure, domain architecture, and the organization of their genomic loci (often referred to as virulence islands). The intricate complexity across various levels has posed a formidable challenge for the identification and characterization of novel toxin/effector systems and their biochemical mechanisms through conventional experimental and bioinformatic approaches. Over the past decade, my research has made significant contribution in unraveling the organizational and diversification principles governing several toxin/effector systems, encompassing both protein domain architecture and genomic loci. Through the integration of these principles into a unique domain- centric guilt-by-association analysis pipeline, I have successfully identified, unified, and categorized an extensive array of known and novel toxin systems into a realm of polymorphic toxin and effector systems (PTSs). This approach has enabled the dissection of hundreds of domains and the prediction of diverse functions and activities for toxins/effectors, immunity proteins, and other associated components. Despite these advancements, the ever-expanding genomic data highlights the presence of many undiscovered systems, emphasizing the need for more sophisticated approaches to fully comprehend these complex molecular mechanisms. In this proposal, my aim is to develop innovative machine learning (ML)-based computational pipelines and resource for the systematic exploration and analysis of complex toxin/effector systems. Additionally, I will leverage our unique analysis strategies to uncover novel toxin systems linked to specific pathogens responsible for severe human and plant diseases, as well as crucial bacterial functions like antiphage mechanisms and species competition.
