Fig. 6: Mtb-reactive metaclones constitute a small proportion of the Day 7 TST repertoire but capture the most public response.

A Abundance of HLA class II-restricted Metaclonotypist β-chain metaclones in TCR repertoires (each down-sampled to 16,000 TCRs; n = 20 Blood, n = 16 Day 2 TST, n = 128 Day 7 TST), shown as percentage of all TCRs or of all expanded ( > 1) TCRs. Boxplots display median and inter-quartile range (IQR), with outlier data points (more than 1.5 × IQR beyond the box hinges). Statistical significance was assessed with unpaired, two-sided Wilcoxon tests and corrected for multiple testing (ns FDR > 0.05, * FDR < 0.05, ** FDR < 0.01, **** FDR < 0.0001). B Abundance of public and private Mtb-reactive CDR3s in the same individual quantified as percentage of all Day 2 or all Day 7 TST TCRs, stratified by clone size ( = TCR count). β−chain TCRs were classified as public if they matched a Metaclonotypist metaclone, or else as private if they matched a private PPD-reactive CDR3 sequence identified from ex vivo stimulated PBMC from the same individual (see Fig. 3). This analysis was restricted to individuals with paired in vitro stimulation experiments (n = 11 Day 2 TST, n = 10 Day 7 TST), and performed on repertoires down-sampled to 16,000 TCRs each. C HLA-class II restricted Metaclonotypist β-chain metaclones (blue), CDR3s with published Mtb reactivity (yellow), and CDR3s present in Day 7 TSTs (red) were each ranked by their publicity across 128 Day 7 TST β repertoires (each down-sampled to 16,000 TCRs) and plotted against the cumulative proportion of participants expressing the TCR. Presence of TCRs was assessed using either all TCR sequences in each sample or only expanded TCRs (present more than once). D Proportion of participants with a cognate HLA contributing to a metaclone in the discovery dataset (n = 151 Day 7 TST) and frequency of the cognate HLA for each HLA class II-restricted Metaclonotypist metaclones represented by individual points.