Purpose: The aim of this research was to assess the application of network theory to the investigation of microbial biofilm (bacterial biofilm co-aggregation networks).
Methods: Network modeling was performed on data from 2 previously published studies on freshwater biofilm systems. Adjacency and connectivity matrices were initially constructed and the following parameters were calculated: clustering co-efficient; characteristic path length; average degree of graph; links per species; connectedness and diameter of graph. The effect of node loss was also determined.
Results: Both biofilm systems showed comparable path lengths and clustering co-efficients above those of associated random graphs. Biofilm networks contain hub organisms and properties resembling those of other ecological networks. Preferential removal of the most highly connected nodes results in a greater number of secondary node loss than random node removal.
Significance: Network modeling may have applicability to biofilm studies. Co-aggregation information may be useful for the development of control strategies as organisms known to co-aggregate with bacteria within a biofilm, but which are detrimental to the biofilm could be introduced to the community. It also identifies 'hub' organisms essential to biofilm formation which can be preferentially targeted. This work suggests that in biofilm communities all the members are close neighbors and that a disinfection protocol that focused on one particular organism may be unnecessary as perturbation effects have the potential to propagate very quickly through an aquatic biofilm system. A focus on the whole biofilm system may still be a preferable option.