ABSTRACT Clostridioides difficile (Cd) infection (CDI) is a leading cause of healthcare-associated infections in the U.S., affecting ~500,000 people and costing ~$5.4B annually. Despite substantial advances in investigating Cd pathogenesis and transmission over the past 20 years, CDI remains a significant public health burden. Conventional antimicrobial therapy often fails to clear Cd, can inflict extensive collateral damage to the commensal gut microbiome (GM), and frequently leads to recurrent CDI, for which treatment options are limited. A major impediment in mitigating Cd-associated disease is current unclear understanding of the biological basis for the broad spectrum of disease severity, from asymptomatic Cd colonization to recurrent fatal CDI. In this context, elucidating mechanisms of Cd colonization and disease necessitates explicit consideration of confounding GM interactions that increase a patient's CDI risk. Accordingly, we propose a quantitative investigation of pathogen-GM-host dynamics of clinically prominent Cd lineages and their contribution to CDI, leveraging a unique cohort of >30,000 Cd-associated patient stools with accompanying well-curated clinical metadata. We hypothesize that high-resolution genomic analyses of Cd strains and pathogen-GM interactions will enable sensitive identification of novel genomic elements in virulent lineages associated with variable CDI outcomes. The rationale for this proposal stems from i) the recognition that most Cd research relies on studying hypervirulent, previously epidemic lineages that are no longer prevalent in the clinic, ii) the need for stable, reproducible GM-humanized animal models of CDI to investigate pathogen and human GM variables, and iii) the observation that taxonomic and functional features of the GM, including bacteriophages, drive Cd's adaptation as a pathobiont. Our central motivation is to expand the representation of prominent Cd lineages and leverage our improved understanding of pathogen-GM dynamics, and our innovative technologies, to enable the design of novel diagnostic, therapeutic, and management avenues to improve patient outcomes. This will be achieved via three aims: 1) investigate pangenome determinants of CDI risk in clinically representative Cd strains, 2) model and predict how different patient GM taxonomic and functional architectures, integrated with clinical metadata, relate to CDI outcomes, and 3) characterize and predict Cd strain-specific dynamics by employing innovative, defined GM-humanized gnotobiotic mouse models of Cd colonization and CDI. Our analyses are significant as they pursue new avenues of precisely investigating pathobiont-commensal variables that influence CDI, a model bacterial-microbiome infectious disease with extensive morbidity and mortality. This proposal is innovative in improving understanding and prediction of Cd disease spectrum via novel complementary technologies including GM-humanized gnotobiotic mouse models to identify...