ECO-EVOLUTIONARY DYNAMICS OF MICROBIAL COMMUNITY ASSEMBLY: COMPETITION, COOPERATION, HIGHER-ORDER AND KEYSTONE INTERACTION
Keywords:
microbial ecology; eco-evolutionary dynamics; community assembly; higher-order interactions; competition; cooperation; cross-feeding; keystone taxa; synthetic microbial communities; priority effects; spatial ecologyAbstract
Microbial communities assemble through reciprocal feedback between ecological interactions and evolutionary change. Classical assembly theory emphasizes selection, dispersal, diversification and ecological drift, but microbial systems can evolve on the same timescales as colonization, competition, facilitation, disturbance and succession. This review develops an explicitly eco-evolutionary framework in which community assembly establishes ecological interactions; those interactions impose selection; genetic and phenotypic adaptation modifies resource use, antagonism, cross-feeding and environmental modification; and the altered interaction network feeds back to subsequent assembly, invasion resistance, stability and ecosystem function. Competition, cooperation and keystone effects are therefore interpreted as dynamic outcomes rather than fixed species properties. Particular emphasis is placed on higher-order interactions, because pairwise measurements can fail when a third species changes metabolite availability, pH, inhibition or spatial access. Recent synthetic-community experiments show that information from three-member combinations can substantially improve bottom-up prediction of multispecies structure. Synthetic microbial communities (SynComs) are highlighted as a central experimental platform for testing interaction mechanisms, keystoneness, spatial organization and evolutionary stability. We also distinguish taxonomic keystones from keystone functions and genes, and critically evaluate co-occurrence networks, centrality, dropout experiments, metabolic modeling, machine learning and longitudinal perturbation as methods for identifying causal community influence. The synthesis argues that predictive microbial ecology will require integration of perturbation experiments, longitudinal multi-omics, spatial measurements, trait evolution and mechanistic modeling. This perspective is relevant to host microbiomes, agriculture, environmental engineering and the rational design of stable microbial consortia.


