OOS 3-10 - An integrated framework for quantifying ecological processes underlying microbial community assembly

Monday, August 7, 2017: 4:40 PM
Portland Blrm 256, Oregon Convention Center
Jizhong Zhou, Institute for Environmental Genomics and Department of Microbiology and Plant Biology, University of Oklahoma, Norman, OK and Daliang Ning, Consolidated Core Laboratory, The University of Oklahoma, Norman, OK

A central issue in microbial ecology is understanding ecological processes controlling community assembly. Ecological processes could be classified into four major categories: selection, dispersal, diversification, and drift. Their relative roles were found obviously vary across different spatial and temporal scales. Recently quantitative approaches and/or frameworks were developed to compare the relative importance of niche-based (deterministic) and neutral processes (stochastic processes). They were based on multivariate correlation analysis, neutral theory model, or null model analysis. However, no framework is currently available to quantify relative importance of all the four major types of processes in shaping microbial community structure. Our objective is to develop an integrated framework to fill this gap. The community structure turnovers governed by selection, dispersal, and diversification were sequentially quantified based on null model analysis of phylogenetic, taxonomic, and molecular β diversity, and the remained should be mainly governed by drift. The probable difference of assembly mechanisms between different phylogeny groups were respected by a binning method based on phylogenetic signal of niche preference. The relative importance of each process in each community turnover was calculated as individual-based percentage of group turnovers governed by the process.


We applied this integrated framework on two continental-scale datasets of soil prokaryotic communities in six forests at North America, and paddy at 13 sites across China. We hypothesized that the influence of heterogeneous selection, dispersal limitation, and in situ (divergent) diversification would increase as geographic distance increases, while homogeneous selection, homogenizing dispersal, and regional (convergent) diversification would be less important as geographic distance increases, and the influence of drift wouldn’t show any clear trend. The results of the integrated framework supported these hypotheses, except very low influence of divergent diversification. Different phylogeny bins showed different mechanisms. For instance, the three most abundant bins, which consisted of OTUs belonging to Rhizobiales, Verrucomicrobia, and Acidobacteria, respectively, were governed by quite different processes in tropic and temperate forest soils. Interestingly, the influence variation of each process should very consistent trends and magnitudes between the microbial communities in American forest soil and China paddy soil, suggesting it could be a general rule of biogeographic pattern of microbial community assembly mechanisms.