Inferred microbial networks lack replicability: consequences for next-generation biomonitoring

Abstract

Microbial interaction networks support many ecosystem services, including the regulation of crop diseases. One of the current challenges is automatically reconstructing these networks from metabarcoding data and monitoring their responses to environmental change. Here, we evaluated the ability of network inference methods to detect changes in crop-associated microbial networks. We used grapevine as a model plant system and assessed the impact of vineyard management (conventional versus organic) on the alpha-and beta-diversity of the fungal communities of grapevine leaves. We also inferred replicated networks of fungal associations, to compare network alpha-and beta-properties between management systems and to generate hypotheses concerning fungus-fungus interactions. We found that the richness, diversity and evenness of fungal communities were significantly higher in organic plots, and that community composition differed between management systems. Erysiphe necator, the causal agent of grapevine powdery mildew, was significantly more abundant in conventional plots, consistent with visual records of disease symptoms, whereas several yeast species were significantly more abundant in organic plots. Vineyard management also had a significant impact on the beta-properties of fungal association networks, but the high turnover of associations between plots precluded the generation of robust hypotheses concerning interactions between fungal taxa, casting doubts on the relationship between microbial association networks and plant health. Network inference methods therefore require improvement and validation before use in the next-generation biomonitoring of disease control services provided by the crop microbiota. As things stand, community-level data appear to be a more reliable and statistically powerful option than network-level data for monitoring the ecosystem services provided by the plant microbiota.

Ecosystems monitoring powered by environmental genomics: A review of current strategies with an implementation roadmap

“…Based on empirical evidence of the variation in network structure under environmental disturbance (Karimi, Meyer, Gilbert, & Bernard, 2016;Ma et al, 2019;Tylianakis, Tscharntke, & Lewis, 2007;Zhou et al, 2011), their properties have been suggested as potential indicators of ecosystem functioning and integrity (Bohan, Caron-Lormier, Muggleton, Raybould, & Tamaddoni-Nezhad, 2011;Bohan et al, 2017;Delmas et al, 2019;Gray et al, 2014;Karimi et al, 2017;Lau, Borrett, Baiser, Gotelli, & Ellison, 2017;Pellissier et al, 2018;Tylianakis & Morris, 2017). In recent years, a growing interest in these approaches has led to a series of studies employing EG to infer ecological networks from microbial community data (Lupatini et al, 2014;Pauvert, Vallance, Delière, Buée, & Vacher, 2019;Pérez-Valera et al, 2017;Zappelini et al, 2015;Zhou et al, 2011) or from macroinvertebrates (Compson et al, 2019), in order to explore the links between network properties such as connectance, centrality or nestedness, and ecosystem functioning. For instance, it has been shown that bacterial communities in anthropized soil may have fewer potentially interacting taxa, than in natural soil (Lupatini et al, 2014).…”

“…While promising, exploring the links between the properties of ecological networks inferred from EG data and ecosystem functioning is still in its infancy (Faust et al., 2012 , 2015 ; Laroche, Wood, et al, 2018 ; Lawes et al., 2017 ; Li et al., 2018 ; Lima‐Mendez et al., 2015 ; Pauvert et al., 2019 ). Multiple methodological issues limit the inference of robust networks from EG data based on co‐occurrences in space or time.…”

“…For example, read counts are strictly compositional, representing relative abundance of the marker itself, rather than presence or absolute abundances (but see Friedman & Alm, 2012 ; Kurtz et al., 2015 ). Further, it is challenging to control for covariates and confounding environmental parameters (but see Chiquet, Mariadassou, & Robin, 2018 ; Cougoul, Bailly, & Wit, 2019 ; Momal, Robin, & Ambroise, 2019 ; Tackmann, Matias Rodrigues, & von Mering, 2019 ; Tamaddoni‐Nezhad et al, 2013 ), replicability of inference (Pauvert et al., 2019 ) and the relative merits of statistical and logical inference (Vacher et al., 2016 ). Robust networks also require considerably more replicates than are typically collected in EG studies, which increase both time and costs.…”

“…Based on empirical evidence of the variation in network structure under environmental disturbance (Tylianakis et al, 2007;Zhou et al, 2011;Karimi et al, 2016;Ma et al, 2018), their properties have been suggested as potential indicators of ecosystem functioning and integrity (Gray et al, 2014;Karimi et al, 2017;Bohan et al, 2011Bohan et al, , 2017Lau et al, 2017;Tylianakis et al, 2017;Pellissier et al, 2018;Delmas et al, 2019). In recent years, a growing interest in these approaches has led to a series of studies employing EG to infer ecological networks from microbial communities data (Zhou et al, 2011;Lupatini et al, 2014;Zappelini et al, 2015;Pérez-Valera et al, 2017;Pauvert et al, 2019) and from macroinvertebrates (Compson et al, 2019) to explore the links between network properties (e.g. connectance, centrality, nestedness) and ecosystem functions.…”

Ecosystems Monitoring Powered by Environmental Genomics: A Review of Current Strategies with An Implementation Roadmap

“…Based on empirical evidence of the variation in network structure under environmental disturbance (Tylianakis et al, 2007;Zhou et al, 2011;Karimi et al, 2016;Ma et al, 2018), their properties have been suggested as potential indicators of ecosystem functioning and integrity (Gray et al, 2014;Karimi et al, 2017;Bohan et al, 2011Bohan et al, , 2017Lau et al, 2017;Tylianakis et al, 2017;Pellissier et al, 2018;Delmas et al, 2019). In recent years, a growing interest in these approaches has led to a series of studies employing EG to infer ecological networks from microbial communities data (Zhou et al, 2011;Lupatini et al, 2014;Zappelini et al, 2015;Pérez-Valera et al, 2017;Pauvert et al, 2019) and from macroinvertebrates (Compson et al, 2019) to explore the links between network properties (e.g.…”