Background: Inflammatory bowel disease (IBD), including Crohn's disease (CD) and ulcerative colitis (UC), is considered to result from the interaction of environmental factors, including intestinal microbiota, with host immune mechanisms in genetically susceptible individuals. Association studies in human and mice indicate that no single organisms are causal for IBD, but instead reflect a broad dysbiosis, suggesting that distinct organisms sharing similar traits may be operative. Owing to the complexity and low inter-individual overlap of the intestinal microbiota, no biologically and ecologically integrated microbial signature of this dysbiosis has been identified in IBD. Results: We performed a study of endoscopic lavage samples from different anatomical sites in 64 subjects (32 normal (NM), 16 CD and 16 UC patients in remission). The V4 region of the 16S ribosomal RNA gene of 190 total mucosal lavage samples was sequenced on an Illumina HiSeq 2000. The microbial composition of lavage specimens was more similar to mucosal biopsies versus fecal compartment (phylum-level Pearson similarity analysis with published datasets), suggesting that mucosal lavage sampling is a good reflection of mucosal-associated microbiota. Phylogenetic diversity was reduced in IBD patients compared to NM subjects. While expected from prior studies of active disease, this is the first demonstration that such differences persist in clinically quiescent phases of disease. A subset of IBD-enriched microbial communities was identified by PCoA of the unweighted UniFrac distance matrix. Network-based analysis of microbial co-occurrence patterns identified 8 functional microbial communities (FMCs). Only one FMC was related to any of the previously reported fecal enterotypes. Bioinformatic validation with two independent mucosal (biopsy) datasets showed that these FMCs are a significantly preserved characteristic of the mucosal surface ecology. Two FMCs were positively and negatively associated with IBD phenotype, and with ATG16L1 and NOD2 risk alleles. Conclusions: We have introduced mucosal lavage as a new sampling method to study host-microbial interaction, and module analysis to identify integrated mucosal ecology parameters for microbial community structure. Our results demonstrate the feasibility of detecting subtle variation between NM and IBD microbiota using relatively short Illumina reads. The identification of FMCs indicated that the intestinal microbiota can be defined as a set of reproducible microbial communities, and that a subset of these are selectively augmented or depleted in IBD patient samples, and with related disease-associated genotypes. Such disease-associated communities provide a new tool to define disease states as microbial ecologic units, and potentially novel integrated units to monitor deleterious microbial states and for therapeutic targeting.