Skip to main content
. 2018 Oct 15;9:2452. doi: 10.3389/fmicb.2018.02452

Table 2.

Statistics for pairwise beta dispersion and PERMANOVA when using different dissimilarity measures on weekly microbiome data.

Beta dispersion Bray-curtis Unweighted unifrac Weighted unifrac
Day 03–07 Day08–14 p = 0.0061142 (**) p = 0.00014712 (***) p = 9.6914e−05 (***)
Day15–24 n.s. p = 0.010418 (*) p = 2.5203e−09 (***)
Day25–35 p = 0.042066 (*) p = 0.00015112 (***) p = 3.5789e−12 (***)
Day08–14 Day15–24 p = 0.00077017 (***) n.s. p = 0.019953 (*)
Day25–35 n.s. n.s. p = 0.0011717 (**)
Day15–24 Day25–35 p = 0.0075651 (**) p = 0.020128 * n.s.
PERMANOVA
Groups R2 = 0.16763 (p = 0.001) (***) R2 = 0.06048 (p = 0.001) (***) R2 = 0.17577 (p = 0.001) (***)
BW_Mean R2 = 0.11721 (p = 0.001) (***) R2 = 0.03964 (p = 0.001) (***) R2 = 0.08723 (p = 0.001) (***)
FI R2 = 0.11856 (p = 0.001) (***) R2 = 0.04069 (p = 0.001) (***) R2 = 0.09301 (p = 0.001) (***)
FCR R2 = 0.1086 (p = 0.001) (***) R2 = 0.03842 (p = 0.001) (***) R2 = 0.11787 (p = 0.001) (***)
Gain R2 = 0.11886 (p = 0.001) (***) R2 = 0.04146 (p = 0.001) (***) R2 = 0.0998 (p = 0.001) (***)

Asterisks denote a statistically significant difference (

*

p < 0.05,

**

p < 0.01,

***

p < 0.001).

In beta dispersion analysis, the pair-wise differences in distances from group center/mean were subjected to ANOVA after performing Principle Coordinate Analysis, and if significant (p ≤ 0.05) the values are shown. In PERMANOVA analysis, R2 represents the proportion of variability explained, for example, using “Groups” and “Bray-Curtis” dissimilarity, the weeks explain 16.8% variability in microbial community structure.