The COVID-19 global pandemic has led to more than 470,000 deaths. This disease is especially perilous for the elderly - 80% of deaths in the US have been individuals over the age of 65, and the social isolation created by lockdowns have increased risks of serious physical and mental health issues. COVID-19 is a heterogeneous disease exhibiting a broad spectrum of symptoms, ranging from mild (e.g. loss of smell, dry cough) to critical (e.g. cytokine storm, renal failure, cardiovascular damage, respiratory failure, lethal blood clotting, neurological disorders). This clinical heterogeneity demands a precision medicine approach that elucidates distinct pathways underlying the disease, develops treatments for each pathway, and defines biomarker patterns to diagnose patients for classification within the subsets. A key benefit of precision medicine is that drugs may be repurposed or may already exist to treat specific subsets of infected individuals. For example, one critical outcome for COVID-19 infection is the onset of a cytokine storm, in which the body's immune system gets caught in a positive feedback loop, leading to shock and rapid failure of multiple organs. There are existing drugs for treating cytokine storm syndrome, but practitioners have no clear guidelines if such treatments are beneficial or destructive. If the individual is not in a hyperinflammatory state, the administration of these drugs could cripple their immune response, leading to increased viral load. Plasma biomarker patterns of proteins and metabolites hold potential to identify impending cytokine storms and other lethal outcomes. To advance precision medicine for COVID-19 treatment, this work will generate large-scale omics data and evaluate levels of proteins and metabolites for plasma drawn from 350 COVID-19 positive cases and 750 normal controls. These data will be immediately released to the research community. Our research team will take a concerted multipronged approach for analyzing these data using diverse complementary techniques. Our labs' research focuses on the discovery of combinations of genes and proteins expressing synchronously and the associations of these combinations with traits of interest, as well as endophenotype discovery. In addition to thorough single analyte analyses, this research will employ three computational strategies to reveal combinations of factors defining patterns: 1) network modeling, 2) explainable-AI systems biology, and 3) linear programming. These intensive analyses will require significant computational resources and we will utilize Summit at Oak Ridge National Laboratory, one of the most powerful supercomputers in the world, for these tasks. The comprehensive protein and metabolite profiles, based on a large cohort of COVID-19 cases and normal controls, along with our rigorous interrogation of these data for complex biomarker patterns indicative of patient outcomes, hold unprecedented potential to drive solid advances in preci...