Main content area

Dominance network analysis provides a new framework for studying the diversity–stability relationship

Ma, Zhanshan (Sam), Ellison, Aaron M.
Ecological monographs 2019 v.89 no.2 pp. e01358
community structure, data collection, etiology, humans, microbiome, models
The diversity–stability relationship is a long‐standing, central focus of community ecology. Two major challenges have impeded studies of the diversity–stability relationship (DSR): the difficulty in obtaining high‐quality longitudinal data sets; and the lack of a general theoretical framework that can encompass the enormous complexity inherent in “diversity,” “stability,” and their many interactions. Metagenomic “Big Data” now provide high quality longitudinal data sets, and the human microbiome project (HMP) offers an unprecedented opportunity to reinvigorate investigations of DSRs. We introduce a new framework for exploring DSRs that has three parts: (1) a cross‐scale measure of dominance with a simple mathematical form that can be applied simultaneously to individual species and entire communities and can be used to construct species dominance networks (SDNs); (2) analysis of SDNs based on special trio motifs, core‐periphery, rich‐club, and nested structures, and high salience skeletons; and (3) a synthesis of coarse‐scale core/periphery/community‐level stability modeling with fine‐scale analysis of SDNs that further reveals the stability properties of the community structures. We apply this new approach to data from the human vaginal microbiome of the HMP, simultaneously illustrating its utility in developing and testing theories of diversity and stability while providing new insights into the underlying ecology and etiology of a human microbiome‐associated disease.