Robust detection of SARS-CoV-2 exposure in the population using T-cell repertoire profiling

Abstract

The COVID-19 pandemic clearly demonstrates the need to monitor the spread of infectious diseases and population immunity. Probing adaptive immunity by sequencing the repertoire of antigen receptors (Rep-Seq) encoding specificity and immunological memory has become a method of choice for immunology studies. Rep-Seq can detect the imprint of past and ongoing infections and study individual responses to SARS-CoV-2 as shown in a number of recent studies. Here we apply a machine learning approach to two large datasets with more than 1200 high-quality repertoires from healthy and COVID-19-convalescent donor repertoires to infer T-cell receptor (TCR) repertoire features that were induced by SARS-CoV-2 exposure. Proper standardization of Rep-Seq batches, access to human leukocyte antigen (HLA) typing and both α- and β-chain sequences of TCRs allowed us to generate a high-quality biomarker database and build a robust and highly accurate classifier for COVID-19 exposure applicable to individual TCR repertoires obtained using different protocols, paving a way to Rep-Seq-based immune status assessment in large cohorts of donors.