Fig. 6: Bact2 enterotype assignment is as stable over time as other enterotypes despite higher between- and within- subject community variation.

a For all enterotype switches, discrepancies between expected and observed switching rates (standardized residuals of a Chi Square goodness of fit test with 10 categories for all possible enterotype switches, P = 2 × 10⁻¹⁶) (y-axis), were set out against community variation (median BC dissimilarity of the constellation using all applicable samples in the study cohort, N = 20, n = 694) (x-axis). b Between-subject community variation per enterotype, calculated as the BC dissimilarity between samples of different individuals of the same enterotype within the FGFP cohort (N = 1103; N = 401,331,186,185 for the Bacteroides1 (B₁), Ruminococcaceae (R), Bacteroides2 (B₂), and Prevotella (P) enterotypes, respectively). Between-subject community variation is significantly higher in Bact2 versus all other enterotypes (ANOVA on multivariate homogeneity of group dispersions on BC dissimilarities between individuals across enterotypes, N = 1103, Q < 10⁻²⁹ for all comparisons with Bact2). c Within-subject community variation per enterotype, calculated as the median BC dissimilarity between all samples of an individual enterotyped as Bact1, Rum, Bact2, or Prev for the majority of their samples (N = 19, excluding participant 808). Within-subject community variation is significantly higher in Bact2 versus Bact1 and Rum enterotypes (Dunn test, two-sided, N = 19, P = 0.0392, and P = 0.0096, respectively). The body of the box plots represents the first and third quartiles of the distribution, and the median line. Whiskers extend from the quartiles to the last data point within 1.5×IQR, with outliers beyond. Significance levels: ***: 0.001, **: 0.01, *: 0.05.