Due to the high sequence similarity to the human genome, the Orangutan genebuild was based on a projection of human gene structures. The projections were made through chained whole genome BLASTz alignments. These projected genes were combined with orangutan-specific proteins, and additional human genes were added using exonerate where the projection was unable to make satisfactory gene models. UTRs were added using orangutan-specific ESTs and cDNAs as well as human cDNAs.