To test which ecological processes may govern bacterial community assembly across groups, we used the methods of Stegen et al. (2013) to calculate the potential ecological processes, including the variable or homogeneous selection, dispersal limitation, homogenizing dispersal, and undominated processes (or called as “ecological drift” in [46]. To quantify community ecological processes, phylogenetic ecological diversity of bacterial communities was computed between any two samples in each group. The package “picante” in R was used to calculate the weighted beta nearest taxon index (β-NTI) parameters. The integration of Bray–Curtis-based Raup–Crick (RCbray) and β-NTI was applied to infer the relative contributions of the above processes dominating the skin microbiota. If the values of β-NTI were >2 or <−2, then this signified that the community turnover was modulated by the variable or homogeneous selection, respectively. In addition, if −2 < β-NTI < 2 as well as RCbray > 0.95 or < −0.95, this indicated that the community composition was regulated by dispersal limitation or homogenizing dispersal, respectively. Lastly, if 2 < β-NTI < 2 and −0.95 < RCbray < 0.95, this indicated that the community diversity was influenced by undominated processes or ecological drift [46]. Variable selection and homogeneous selection belonged to deterministic processes, yet dispersal (namely, dispersal limitation and homogenizing dispersal) processes and drift belong to stochastic processes [32]. Thus, the stochastic and deterministic processes of each group were also calculated.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.