The modENCODE Transcriptome group clustered genes into one of 20 co-expression clusters based on the similarity of their expression profiles throughout development (FBrf0213980). This clustering analysis was based upon FlyBase gene annotations from November 2009 (release Dmel 5.23, FB2009_10). Of the 13,531 genes listed, 11,653 could be assigned unambiguously to a single FlyBase gene record. Cluster assignments for gene IDs with no clear correspondence to a FlyBase gene record were not included. The genes cited in these 11,653 cluster assignments were reviewed for potential conflicts or confounding effects resulting from changes to annotated gene models between Dmel releases 5.23 and 5.38 (June 2011, FB2011_06). In some cases, 2 synonyms of the same gene were assigned to different clusters: these cluster assignments were discarded. Cluster assignments were not incorporated in cases where two gene models assigned to different co-expression clusters were merged in subsequent FlyBase releases. Cluster assignments were not incorporated in cases where two gene models were merged and one of them was not assigned to any co-expression cluster. Cluster assignments were accepted if two gene models from the same cluster were merged, or if a clustered annotated gene was subsequently merged to an unannotated gene (inferred from genetic data only). 184 assignments were discarded. In 8 cases, two genes from the same cluster were merged since FlyBase Dmel release 5.23, so these 16 assignments were mapped to the 8 resulting genes. In all, 11,461 distinct gene-cluster assignments have been incorporated.