Microbial communities are essential for human and environmental health, often forming complex interaction networks responsible for driving ecosystem processes affecting their local environment and their hosts. Disturbances of these communities can lead to loss of interactions and thereby important ecosystem functionality. The research on what drives interactions in microbial communities is still in its infancy, and much information has been gained from the study of model communities. One purpose of using these model microbial communities is that they can be cultured under controlled conditions. Yet, it is not well known how fluctuations of abiotic factors such as temperature affect their interaction networks. In this work, we have studied the effect of temperature on interactions between the members of the model community THOR, which consists of three bacterial species: Pseudomonas koreensis, Flavobacterium johnsoniae, and Bacillus cereus. Our results show that the community-intrinsic properties resulting from their interspecies interactions are highly dependent on incubation temperature. We also found that THOR biofilms had remarkably different abundances of their members when grown at 11, 18, and 25°C. The results suggest that the sensitivity of community interactions to changes in temperature is influenced, but not completely dictated, by different growth rates of the individual members at different temperatures. Our findings likely extend to other microbial communities and environmental parameters. Thus, temperature could affect community stability and may influence diverse processes including soil productivity, bioprocessing, and disease suppression. Moreover, to establish reproducibility between laboratories working with microbial model communities, it is crucial to ensure experimental stability, including carefully managed temperature conditions.

Summary Sequencing of transposon insertion libraries is used to determine the relative fitness of individual mutants at a large scale. However, there is a lack of tools for specifically analyzing data from such experiments with paired sample designs. Here, we introduce CAFE—Coefficient-based Analysis of Fitness by read Enrichment—a software package that can analyze data from paired transposon mutant sequencing experiments, generate fitness coefficients for each gene and condition, and perform appropriate statistical testing on these fitness coefficients. Availability and Implementation CAFE is implemented in Perl and R. The source code is freely available for download under the MIT License from https://github.com/bengtssonpalme/cafe and http://microbiology.se/software/cafe/ Supplementary information Supplementary data are available at Bioinformatics online. The evaluation data can be obtained from https://microbiology.se/sw/cafe/example_data.tgz