More than three hundred genomes have been sequenced in the last decade. A new self-trained gene prediction program that can be applied to any eukaryotic genome will be developed. The method builds the required training set in three steps, beginning with a seed set from existing resources for transcribed regions. The set is extended and combined with a structure model into a complete gene finder. The performance will be evaluated on the reference organism, Toxoplasma gondii, which has both draft sequence and EST sequences. The eukaryotic genomes provide information that will be of benefit in both economic and health considerations. As the coupling between different disciplines grows strong, a more interdisciplinary educational structure will be needed. The education aspects of this effort will address this at both undergraduate and graduate levels.
