ANTIBODIES TO SARS-COV-2: AN EXPLORATORY ANALYSIS CARRIED OUT THROUGH THE BAYESIAN PROFILE REGRESSION

In this paper we aim at characterizing the immune response to SARSCoV- 2 in a cohort study of individuals who received the first dose of of mRna vaccines Pzifer or AstraZeneca. To examine some covariate-related effects on anti-S1 spike IgG levels we adopted a statistical technique known as Bayesian Profile Regression (BPR). The BPR explores the link between a response variable and a set of associated covariate data through cluster membership and supervises the clustering assignment in a unified fashion. In our study, this methodology allowed us to identify three clusters, differentiated according to the antibody titer of respondents, and draw the profile of subjects whose amount of antibodies produced is significantly higher.