The aim of the project is to discover the fundamental principles governing the adaptation to the disturbance of multi-species adherent bacterial communities on a 4-species (4S) bacterial biofilm model of natural origin. In nature, these systems play a crucial role in the biogeochemical cycles of carbon, nitrogen and water. Their disruption can only be accompanied by major consequences on a global scale. However, how these communities respond to climate change is currently unknown. Consideration of this issue at the level of natural ecosystems is difficult to achieve because environmental conditions can not be varied and controlled rationally. On the other hand, bacterial biofilm models in the laboratory are mostly mono-species whereas we are beginning to understand that inter-species interactions are crucial. To better understand the factors that support the adaptation of these communities to disturbances, we propose to bring together biophysicists and microbiologists who will examine the global and molecular responses of the 4S model community to controlled environmental changes in the laboratory. In a first phase, we will try to establish a quantitative phenotypic description coupled with a genetic description of the biofilm 4S installed on a microfluidic platform that will control all the physical and chemical conditions as well as follow in real-time the community. A combinatorial approach - all biofilms from mono- to 4-species will be studied in parallel - will identify inter-species interactions in a reference state. We will then carry out a second phase of a series of controlled perturbations of various chemical, physical and social natures in order to detect characteristic adaptive trajectories (resistance, resilience or redundancy) and to select remarkable adaptation stages - climax or plateau - which will then be studied from the genetic point of view in the 3rd phase of the project. In this study, the transcriptional changes and genomic alterations occurring at the salient points of the selected adaptive trajectories will be studied. This program involves several methodological and technological challenges, such as the implementation of a quantitative description of the phenotype, the automation of the experiment required by the combinatorial approach and the screening of disturbances. The genetic analyzes that will have to be carried out in the multi-species context will require the implementation of the latest technical advances in the field. Our approach aims to link phenotypic and genetic information by pre-selection of relevant trajectories and correlated phenotypic and genetic analyzes on defined points of the trajectory. The completion of our program should first clarify the role of inter-species interactions in the architecture of the mixed community and its capacity to adapt to a given constraint, new experimental tools to analyze adherent bacterial communities, New strategies to control bacterial biofilms, and even new avenues for the artificial assemblage of artificial communities with defined functions. Finally, our work will evaluate the potential of simplified multi-species model systems cultivated in microfluidic geometry to understand and predict the dynamics of natural communities