Fig. 1

CMV serostatus and HLA-I genotype are associated with TCR repertoire diversity. a Variation in number of unique CDR3s and Shannon entropy, two measures of TCR repertoire diversity, across the cohort. b Association of CMV seropositivity (CMV+) with reduced TCR repertoire diversity (Shannon entropy). P = 6.43e-14, two-sided Wilcoxon test. c Association of CMV seropositivity (CMV+) with reduced TCR repertoire diversity (number of unique CDR3s). P = 0.07, two-sided Wilcoxon test. d Association of HLA-I polymorphism with increased number of unique CDR3s in CMV- individuals; HLA-I P = 0.02, estimate = 18,787.8; age P = 0.002, estimate − 1326.3. P-values are from a linear model incorporating number of unique HLA-I alleles and age. e Association of full HLA-I heterozygosity (6 different HLA-I alleles) with number of unique CDR3s in CMV- individuals; full HLA-I heterozygosity P = 0.02, estimate = 29,248.4; age P = 0.002, estimate = − 1342.6. P-values are from a linear model incorporating a binary variable encoding full HLA-I heterozygosity, and age as a continuous variable. f No association between HLA-II polymorphism and number of unique CDR3s in CMV- individuals; HLA-II P = 0.82, estimate = 1224.9; age P = 0.006, estimate = − 1182.1. P-values are from in a linear model incorporating number of unique HLA-II alleles and age. g No association between full HLA-II heterozygosity (10 unique HLA-II alleles) and number of unique CDR3s in CMV- individuals; HLA-II P = 0.21, estimate = − 17,362.9; age P = 0.006, estimate = − 1153.4. P-values are from a linear model incorporating a binary variable encoding full HLA-II heterozygosity, and age as a continuous variable