Microbial community-level statistics. (a) Nares samples clustered using principal coordinates analysis (PCoA) of the θ 'similarity' coefficient, which calculates similarity between two samples based on i) number of species in common between two samples and ii) their relative abundances. Colors represent different Tanner stages. Samples that have similar principal coordinates appear closer together, that is, are more similar. Ellipses represent the 95% confidence intervals based on standard deviation. Biplot arrows indicate the five most significant unique consensus taxonomies contributing to variation along axis 1. (Spearman correlations with axes and associated P-values are shown in Table S3 in Additional file 1). a, Propionibacterium; b, Corynebacterium; c, Streptococcus; d, Turicella; e, Anaerococcus. Additional skin sites are shown in Figure S2 in Additional file 2. (b) Heatmap of mean θ similarity coefficients of pairwise comparisons of communities within or between individuals of Tanner stage 1 and Tanner stage 5 (within-group, Tanner 1 versus Tanner 1 or Tanner 5 versus Tanner 5; between-group, Tanner 1 versus Tanner 5). Darker color indicates higher similarity: θ = 0 indicates that two samples have no species in common and θ = 1 indicates that two samples are identical. Site abbreviations: Af, antecubital fossa; N, nares; Pf, popliteal fossa; Vf, volar forearm.