BackgroundGrowth rates, interactions between community members, stochasticity, and immigration are important drivers of microbial community dynamics. In sequencing data analysis, such as network construction and community model parameterization, we make implicit assumptions about the nature of these drivers and thereby restrict model outcome. Despite apparent risk of methodological bias, the validity of the assumptions is rarely tested, as comprehensive procedures are lacking. Here, we propose a classification scheme to determine the processes that gave rise to the observed time series and to enable better model selection.ResultsWe implemented a three-step classification scheme in R that first determines whether dependence between successive time steps (temporal structure) is present in the time series and then assesses with a recently developed neutrality test whether interactions between species are required for the dynamics. If the first and second tests confirm the presence of temporal structure and interactions, then parameters for interaction models are estimated. To quantify the importance of temporal structure, we compute the noise-type profile of the community, which ranges from black in case of strong dependency to white in the absence of any dependency. We applied this scheme to simulated time series generated with the Dirichlet-multinomial (DM) distribution, Hubbell’s neutral model, the generalized Lotka-Volterra model and its discrete variant (the Ricker model), and a self-organized instability model, as well as to human stool microbiota time series. The noise-type profiles for all but DM data clearly indicated distinctive structures. The neutrality test correctly classified all but DM and neutral time series as non-neutral. The procedure reliably identified time series for which interaction inference was suitable. Both tests were required, as we demonstrated that all structured time series, including those generated with the neutral model, achieved a moderate to high goodness of fit to the Ricker model.ConclusionsWe present a fast and robust scheme to classify community structure and to assess the prevalence of interactions directly from microbial time series data. The procedure not only serves to determine ecological drivers of microbial dynamics, but also to guide selection of appropriate community models for prediction and follow-up analysis.Electronic supplementary materialThe online version of this article (10.1186/s40168-018-0496-2) contains supplementary material, which is available to authorized users.

Western diet (WD) is one of the major culprits of metabolic disease including type 2 diabetes (T2D) with gut microbiota playing an important role in modulating effects of the diet. Herein, we use a data-driven approach (Transkingdom Network analysis) to model host-microbiome interactions under WD to infer which members of microbiota contribute to the altered host metabolism. Interrogation of this network pointed to taxa with potential beneficial or harmful effects on host’s metabolism. We then validate the functional role of the predicted bacteria in regulating metabolism and show that they act via different host pathways. Our gene expression and electron microscopy studies show that two species from Lactobacillus genus act upon mitochondria in the liver leading to the improvement of lipid metabolism. Metabolomics analyses revealed that reduced glutathione may mediate these effects. Our study identifies potential probiotic strains for T2D and provides important insights into mechanisms of their action.

The human microbiome has been implicated in affecting health outcomes in premature infants, but the ecological processes governing early life microbiome assembly remain poorly understood. Here, we investigated microbial community assembly and dynamics in extremely low birth weight infants (ELBWI) over the first 2 weeks of life. We profiled the gut, oral cavity and skin microbiomes over time using 16S rRNA gene amplicon sequencing and evaluated the ecological forces shaping these microbiomes. Though microbiomes at all three body sites were characterized by compositional instability over time and had low body-site specificity (PERMANOVA, r2 = 0.09, p = 0.001), they could nonetheless be clustered into four discrete community states. Despite the volatility of these communities, deterministic assembly processes were detectable in this period of initial microbial colonization. To further explore these deterministic dynamics, we developed a probabilistic approach in which we modeled microbiome state transitions in each ELBWI as a Markov process, or a “memoryless” shift, from one community state to another. This analysis revealed that microbiomes from different body sites had distinctive dynamics as well as characteristic equilibrium frequencies. Time-resolved microbiome sampling of premature infants may help to refine and inform clinical practices. Additionally, this work provides an analysis framework for microbial community dynamics based on Markov modeling that can facilitate new insights, not only into neonatal microbiomes but also other human-associated or environmental microbiomes.