Publication Details

Category Text Publication
Reference Category Journals
DOI 10.3390/microorganisms8020190
Licence creative commons licence
Title (Primary) Keystone species and modularity in microbial hydrocarbon degradation uncovered by network analysis and association rule mining
Author Centler, F.; Günnigmann, S.; Fetzer, I.; Wendeberg, A.
Source Titel Microorganisms
Year 2020
Department UMB
Volume 8
Issue 2
Page From art. 190
Language englisch
Supplements https://www.mdpi.com/2076-2607/8/2/190/s1
Keywords hydrocarbon degradation; microbial communities; spatial scales; sample resolution; co-occurrence analysis; association rule mining; network analysis
Abstract Natural microbial communities in soils are highly diverse, allowing for rich networks of microbial interactions to unfold. Identifying key players in these networks is difficult as the distribution of microbial diversity at the local scale is typically non-uniform, and is the outcome of both abiotic environmental factors and microbial interactions. Here, using spatially resolved microbial presence-absence data along an aquifer transect contaminated with hydrocarbons, we combined co-occurrence analysis with association rule mining to identify potential keystone species along the hydrocarbon degradation process. Derived co-occurrence networks were found to be of a modular structure, with modules being associated with specific spatial locations and metabolic activity along the contamination plume. Association rules identify species that never occur without another, hence identifying potential one-sided cross-feeding relationships. We find that hub nodes in the rule network appearing in many rules as targets qualify as potential keystone species that catalyze critical transformation steps and are able to interact with varying partners. By contrasting analysis based on data derived from bulk samples and individual soil particles, we highlight the importance of spatial sample resolution. While individual inferred interactions are hypothetical in nature, requiring experimental verification, the observed global network patterns provide a unique first glimpse at the complex interaction networks at work in the microbial world.
Persistent UFZ Identifier https://www.ufz.de/index.php?en=20939&ufzPublicationIdentifier=22745
Centler, F., Günnigmann, S., Fetzer, I., Wendeberg, A. (2020):
Keystone species and modularity in microbial hydrocarbon degradation uncovered by network analysis and association rule mining
Microorganisms 8 (2), art. 190 10.3390/microorganisms8020190