TABLE 1 .
Variableb | Bray-Curtis dissimilarity |
Weighted Unifrac distance |
||||
---|---|---|---|---|---|---|
F value | R2 | Pr(>F) | F value | R2 | Pr(>F) | |
1st level | ||||||
Host species identity | 4.97 | 19.38 | 0.001 | 5.03 | 19.73 | 0.001 |
Urban intensity | 4.66 | 6.05 | 0.001 | 5.07 | 6.63 | 0.001 |
Tree isolation | 2.78 | 1.81 | 0.001 | 2.92 | 1.91 | 0.009 |
2nd level | ||||||
Urban intensity* | ||||||
Tree isolation | 1.33 | 1.73 | 0.056 | NS | NS | NS |
Host species* | ||||||
Tree isolation | 1.74 | 6.77 | 0.001 | 1.77 | 6.96 | 0.004 |
Urban intensity* | ||||||
Host species | 1.55 | 12.10 | 0.001 | 1.49 | 11.71 | 0.005 |
3rd level | ||||||
Urban intensity* | ||||||
Host species* | 1.23 | 7.98 | 0.017 | NS | NS | NS |
Tree isolation |
Data represent results of PERMANOVA analysis of Bray-Curtis dissimilarities and weighted Unifrac distances. The models explained, respectively, 56% and 47% of the variation in bacterial taxonomical and phylogenetic community structure. Pr(>F), P value; NS, not significant.
An asterisk represents interaction between variables.