Understanding humoral responses to SARS-CoV-2 is critical for improving diagnostics, therapeutics, and vaccines. Deep serological profiling of 232 COVID-19 patients and 190 pre-COVID-19 era controls using VirScan revealed over 800 epitopes in the SARS-CoV-2 proteome, including 10 epitopes likely recognized by neutralizing antibodies. Pre-existing antibodies in controls recognized SARS-CoV-2 ORF1, while only COVID-19 patients primarily recognized spike and nucleoprotein. A machine learning model trained on VirScan data predicted SARS-CoV-2 exposure history with 99% sensitivity and 98% specificity; a rapid Luminex-based diagnostic was developed from the most discriminatory SARS-CoV-2 peptides. Individuals with more severe COVID-19 exhibited stronger and broader SARS-CoV-2 responses, weaker antibody responses to prior infections, and higher incidence of CMV and HSV-1, possibly influenced by demographic covariates. Among hospitalized patients, males make greater SARS-CoV-2 antibody responses than females.

The human virome plays important roles in health and immunity. However, current methods for detecting viral infections and antiviral responses have limited throughput and coverage. Here, we present VirScan, a high-throughput method to comprehensively analyze antiviral antibodies using immunoprecipitation and massively parallel DNA sequencing of a bacteriophage library displaying proteome-wide peptides from all human viruses. We assayed over 108 antibody-peptide interactions in 569 humans across four continents, nearly doubling the number of previously established viral epitopes. We detected antibodies to an average of 10 viral species per person and 84 species in at least two individuals. Although rates of specific virus exposure were heterogeneous across populations, antibody responses targeted strikingly conserved “public epitopes” for each virus, suggesting that they may elicit highly similar antibodies. VirScan is a powerful approach for studying interactions between the virome and the immune system.