Fig 1.
Workflow for candidate biomarker selection.
Microbial species and molecular feature details: step 1—data stocktaking at the microbial species and biological function level; step 2—assignment of genes to biological function terms. Interaction data and networks: step 3—all-against-all ortholog mapping; step 4—aggregation of protein interaction data; step 5—deriving a molecular process model. Biomarker candidates and experimental evaluation: step 6—selection of biomarker candidates utilizing network topology characteristics; step 7—experimental evaluation of biomarker candidates.
Table 1.
Listing of copper sites.
Table 2.
Microbial species annotation details.
Table 3.
Molecular function terms and COG term assignment.
Fig 2.
Nodes represent protein coding genes and edges resemble aggregate interaction information. The layout reflects node degree, the node diameter scales with stress centrality. Node color-coding indicates microbial species coverage scaling from one to 25 (upper semicircle, scaling from light to dark blue), and selected COG term assignment in the context of biofilm formation (lower semicircle, orange: cell wall/membrane/envelop biogenesis; green: cell motility; magenta: extracellular structures). Node annotation refers to biomarker candidates according to Table 4.
Fig 3.
Node details of the biofilm process network in relation to experimental data.
Violin plots for (A) degree and (B) stress centrality of network nodes comparing the wild type microbial community situation with laboratory conditions on fold change significance, complemented by correlation of species coverage with (C) degree and (D) stress centrality for nodes holding a significant fold change.
Table 4.
Biomarker candidate panel for capturing the biofilm molecular process.