Variation in the microbiome in response to parasite, bat host, and environment
Ordination and PERMANOVA of microbiome communities provide evidence that aspects of the parasite (i.e., parasite family and species), the host bat (i.e., bat family, bat sex, and bat individual), and landscape factors (i.e., region and sampling site) significantly contributed to bat fly microbiome variation (Table 2; Figure 3 and Figure S5). Other variables significantly contributed to microbiome community differentiation (i.e., bat feeding guild, bat species, protection status of sampling site, habitat fragment area, and isolation), but violated the assumption of homoscedasticity. Parasite species, parasite family, bat feeding guild, bat family, bat species, and sampling site had the largest effect sizes, however many of these variables are correlated with each other.
We used sequential sum of squares with free permutation to account for the hierarchical structure of the study system, and found that parasite species significantly impacted microbiome community structure (Table 3). Habitat fragment area, distance to source, and protection status also significantly contributed to microbiome variation, but violated the assumption of homoscedasticity. As the order of variables in sequential sum of squares can impact their significance when sample sizes are uneven, we also examined the impact of parasite, host bat, and landscape variables on the four most well-sampled species (Table S1; Figure 4). None of the test variables significantly explained microbiome variation without violating PERMANOVA assumptions.