Peptide sequences of Atlantic cod

Transcripts were annotated with UniProtKB, Zebrafish ENSEMBL gene model and Medaka ENSEMBL gene model. The transcripts were grouped in clusters based on similarity. We chose the transcript with the longest match to the genomes, and the transcript with the best annotation BLAST score. In cases where the same transcript had the longest match to the genome and the best annotation blast score, only one transcript was added. As the transcripts are assembled from many samples we do not know if the differences between transcripts in a cluster are splice variants or assembly errors, as there may be errors in the assemblies based on RNA-Seq.