The honey bee (Apis mellifera) genome, sequenced by the Baylor College of Medicine Human Genome Sequencing Center (BCM-HGSC), provides a new resource with which to further develop the honey bee model for sociogenomics, the study of the molecular basis of social life (Robinson G.E. et al. 2005. Nature Rev. Genet. 6:257-270). The long-term objective of BeeBase is to promote the honey bee model system. The first step was to support the community genome annotation effort led by BCM-HGSC. We collaborated with BCM-HGSC and Aaron Mackey at University of Pennsylvania to evaluate and merge five gene prediction sets into a consensus gene set. Consensus gene models were assigned identifiers at BeeBase following the naming system used at FlyBase. The consensus set became the first release of the official honey bee gene set, and served as the starting point for manual annotation by the honey bee research community. Community members worked in groups divided according to the following themes: 1) innate immunity, 2) pesticides and stress resistance, 3) neurobiology and behavior, 4) gene regulation, 5) development and metabolism, and 6) reproduction. They used tools at BeeBase and elsewhere to manually annotate gene models, and they submitted models to the BCM-HGSC annotation database. BeeBase provided BLAST and PSI-BLAST servers, access to all predicted gene sets, and genome browsers featuring tracks for predicted gene sets, homologs, EST, markers, SNP, repeats, and GC content analysis. Community annotation data was transferred from the BCM-HGSC annotation database to BeeBase for identifier assignment. We incorporated over 3000 manually annotated gene-models into release 2 of the official honey bee gene set. The next steps include developing gene pages, annotating gene products with Gene Ontology, incorporating large-scale expression datasets from the BeeSpace Project (University of Illinois-Urbana Champaign), and connecting honey bee gene annotations to the BeeSpace conceptual literature navigation system.