Fig. 2

Age and sex impact on total and antigen-specific antibody levels. a Relationships between Log10-transformed IgG (upper left), IgA (upper right), IgM (bottom left), and IgE (bottom right) levels and age. Regression lines were fitted using linear regression, with Log10-transformed total antibody levels as response variable, and age and sex as treatment variables. Indicated adj. P were obtained using the mixed model and corrected for multiple testing using the FDR method. b, c Effect sizes of significant associations (adjusted P values (adj. P < 0.05) between age (b) and sex (c) on Log10-transformed antigen-specific IgG levels in the 1000 healthy individuals from the Milieu Intérieur cohort. Because of low number of seropositive donors (n = 15), HBc serology was removed from this analysis. Effect sizes were estimated in a linear mixed model, with Log10-transformed antigen-specific IgG levels as response variables, and age and sex as treatment variables. All results from this analysis are provided in Additional file 1: Table S5. Dots represent the mean of the beta. Lines represent the 95% confidence intervals