The gene set for panda was built using the Ensembl pipeline. Gene models are based on genewise alignments of mammal proteins as well as genetically distance proteins from other species, including most vertebrate proteins from Uniprot. Additionally, dog and human proteins were aligned using exonerate.
Potentially missing predictions and partial gene predictions were identified by examining the orthologs, which were then used to build new gene models.