ABSTRACT. The acute respiratory distress syndrome (ARDS) is associated with high mortality, morbidity, and health care costs (Bellani 2016). Dozens of candidate drugs for ARDS have been identified in preclinical models, but none consistently reduced mortality in randomized controlled trials (RCTs). This dismal record is likely driven in part by the heterogenous biology encompassed within the definition of ARDS (Rubenfeld 2015). Unsupervised clustering of plasma biomarkers and clinical variables recently identified two molecular phenotypes of ARDS (Calfee 2014) which may enable predictive enrichment in future RCTs. Uncertainty about the key biologic mechanisms that distinguish these two phenotypes from each other remains a critical knowledge gap. The overall objective of this proposal is to recruit patients to an established cohort of mechanically ventilated patients and identify distinct mechanisms of lung injury in ARDS molecular phenotypes. The central hypothesis of this proposal is that ARDS phenotypes are driven by different dysregulated pathways that result in distinct clinical trajectories and responses to treatment. In Aim 1, I will use single cell RNA sequencing to study tracheal aspirates and develop a model of cell signaling in the lung of each molecular phenotype. I hypothesize hyperinflammatory ARDS is associated with increased Type 1 T-cell polarization and a diminished response to interferons in macrophages. In Aim 2, I will use metatranscriptomic sequencing to characterize differences in the tracheal aspirate microbiome in each phenotype. I hypothesize TA metatranscriptomes will have distinct microbial community composition in each phenotype, which will be characterized by increased burden of enteric bacteria in the hypoinflammatory phenotype and an increased burden of fungi in the hyperinflammatory phenotype. In Aim 3, I will collect longitudinal tracheal aspirate and plasma samples to study the evolution of pro-inflammatory, pro-resolution, and pro-fibrotic pathways in each ARDS phenotype. I hypothesize ARDS phenotypes have distinct trajectories of inflammation and repair pathways in the first week of mechanical ventilation. I will address a critical gap in knowledge required to develop phenotype-specific precision treatments. This K23 award is sponsored by Dr. Carolyn Calfee, an experienced ARDS researcher whose group has pioneered analyses of ARDS molecular phenotypes, and Dr. Stephanie Christenson, a computational biologist with expertise in the transcriptomics of airway diseases. Their mentorship and the research and training plan in this K23 will support my continued career development and allow me to learn essential skills I require to be an independent investigator, including advanced computational analyses, epidemiological methods, and management of a research cohort. Developing these skills will be essential to achieve my long-term goal of understanding the mechanistic pathways distinguishing ARDS phenotypes to identify ...