PT  - JOURNAL ARTICLE
AU  - E.K. Hall
AU  - E.S Bernhardt
AU  - R.L. Bier
AU  - M.A. Bradford
AU  - C.M. Boot
AU  - J.B. Cotner
AU  - P.A. del Giorgio
AU  - S.E. Evans
AU  - E.B. Graham
AU  - S.E. Jones
AU  - J.T. Lennon
AU  - K.J. Locey
AU  - D. Nemergut
AU  - B.B. Osborne
AU  - J.D. Rocca
AU  - J.S. Schimel
AU  - M.P. Waldrop
AU  - M.W. Wallenstein
TI  - Understanding How Microbiomes Influence The Systems They Inhabit: Moving From A Correlative To A Causal Research Framework
AID  - 10.1101/065128
DP  - 2018 Jan 01
TA  - bioRxiv
PG  - 065128
4099  - http://biorxiv.org/content/early/2018/05/24/065128.short
4100  - http://biorxiv.org/content/early/2018/05/24/065128.full
AB  - Translating the ever-increasing wealth of information on microbiomes (environment, host, or built environment) to advance the understanding of system-level processes is proving to be an exceptional research challenge. One reason for this challenge is that relationships between characteristics of microbiomes and the system-level processes they influence are often evaluated in the absence of a robust conceptual framework and reported without elucidating the underlying causal mechanisms. The reliance on correlative approaches limits the potential to expand the inference of a single relationship to additional systems and advance the field. We propose that research focused on how microbiomes influence the systems they inhabit should work within a common framework and target known microbial processes that contribute to the system-level processes of interest. Here we identify three distinct categories of microbiome characteristics (microbial processes, microbial community properties, and microbial membership) and propose a framework to empirically link each of these categories to each other and the broader system level processes they affect. We posit that it is particularly important to distinguish microbial community properties that can be predicted from constituent taxa (community aggregated traits) from and those properties that are currently unable to be predicted from constituent taxa (emergent properties). Existing methods in microbial ecology can be applied to more explicitly elucidate properties within each of these categories and connect these three categories of microbial characteristics with each other. We view this proposed framework, gleaned from a breadth of research on environmental microbiomes and ecosystem processes, as a promising pathway with the potential to advance discovery and understanding across a broad range of microbiome science.ContributionAll listed authors have contributed to the conceptualization, writing, and preparation of the current manuscript.