Generic Model Organism Database project.

The Daphnia pulex genome is rich in tandem duplicate genes, some 20 % of its 30,000+ genes. However some gene predictors have missed or incorrectly located almost half of these. Estimates from genome-wide tile expression suggest an additional 5,000 genes have been missed. Gene prediction for new genomes such as this first crustacean is still an uncertain task. Even in clades with a well-characterized model such as Drosophila, gene finding remains an uncertain task. Prediction tools are increasingly sophisticated and accurate. Today's methods draw on the range of available gene evidence and improved modeling of gene structures. Yet they are sensitive to available gene data and expected structures. They find well-known genes, but fail at accurate detection of novel and diverged genes. Measures from gene duplication and genome-wide tile expression can more accurately locate those genes missed by other methods. Computational methods are being developed to turn these signals to accurate gene models. Application of these methods to arthropod genomes, including Daphnia and Drosophila, uncovers some 10 % to 25% additional species specific and diverged genes. This work includes development of new automated genome analysis pipelines on NSF TeraGrid shared cyberinfrastructure, as part of th