This project seeks to determine associations of COVID19 disease severity with HLA genotypes by developing composite genetic risk scores and to interpret these results in terms of the pattern of particular SARS-CoV-2 encoded peptides preferentially bound by positively and negatively associated allotypes. Our working hypothesis is: individuals with allotypes that bind few peptides encoded by the SARS-CoV-2 genome consequentially have a limited repertoire of T cells capable of recognizing these peptides. Furthermore, because they see only a fraction of the SARS-CoV-2 ‘peptidome’, they will make an impoverished immune response to SARS-CoV-2, resulting in progression to severe disease. We will use high resolution HLA genotyping of patient DNA samples from the Columbia Covid-19-Biobank in conjunction with the linked clinical data to determine the associations of COVID19 disease severity with HLA genotype and develop a composite genetic risk score that would identify those more likely to have a poor outcome, potentially highlighting their need for more specialized immune interventions, and of relevance to vaccination outcomes. Data on peptide binding preferences of the HLA allotypes comprising the genetic risk scores will be used to generate topologic maps showing the differences in the ability of these individuals to recognize features of the SARS-CoV-2 peptidome.