PROJECT SUMMARY Pulmonary fibrosis (PF) represents a global clinical burden that affects over 5,000,000 people and can occur as a result of chemical injury, chronic conditions such as systemic sclerosis, or respiratory infections such as influenza and COVID-19. Commonly, PF has no clinically determinable cause and is diagnosed as idiopathic PF, which presents a median survival time of only 2-4 years after diagnosis. Given the often unclear pathogenesis of PF, there exists a strong clinical need to elucidate the biological mechanisms that contribute to its onset and progression. Nevertheless, the factors that drive spatial heterogeneity and temporal progression in fibrotic architecture are not well understood. Furthermore, the post-fibrotic resolution of aberrant PF matrix remains an elusive goal, for which no single-cell characterizations have been performed to date. Thus, this project aims to establish a spatiotemporal atlas of PF progression that links multi-omics with spatially defined tissue neighborhoods and temporally defined architectural states of fibrosis and post-fibrotic resolution. Mesenchymal cell populations play a critical role in the fibrosis of all major organs, with a number of macrophage and fibroblast subtypes often implicated as mediators of fibrotic ECM deposition. Based on prior studies, I hypothesize that transcriptionally defined macrophage and fibroblast subtypes act as both spatial and temporal nodes of fibrosis and post-fibrotic resolution. To investigate this hypothesis, this project will establish a novel computational atlas of transcriptional/epigenetic cell subtypes, interaction networks, and ultrastructural states that mediate the pathological progression of PF and post-fibrotic repair in mice. Specific Aim 1 will investigate the roles of transcriptionally defined cell subpopulations in the temporal progression and resolution of fibrotic pulmonary architecture, using high-throughput multi-omics (transcriptomic, epigenomic, and ultrastructural) and computational modeling of biological variations over time. Specific Aim 2 will define the spatial tissue neighborhoods of cell- and matrix- mediated interactions in pulmonary fibrosis, by integrating Visium spatial transcriptomics, imputed spatial epigenomics in BABEL, and ultrastructural quantification on consecutive histological slices. Specific Aim 3 will develop a machine learning algorithm for prognosis of clinical outcomes in human pulmonary fibrosis, by unifying histopathological architecture, protein and cell spatial networks, and clinical metadata. Ultimately, this project will establish a multi-omic, cross-species, and computationally rigorous atlas of PF progression and repair that identifies biologically conserved mechanistic pathways and clinically relevant targets for prognosis and therapeutic development.