HiSeq2000 Illumina pair-end sequenced transcript data originating from a pool of four different tissues of a wild-type American mink yielded approximately 90 Gb of raw data. Subsequently, unique contigs were assembled by a combined approach using velvet and phrap. Of these assembled contigs, about 136 000 match the dog genome and nearly 30 000 contigs match the human transcriptome at more than 17 000 unique gene locations. Gene annotation for these contigs was performed employing custom-made scripts run in combination with comparative sequence similarity search and alignment in the dog and human genome using blast algorithms. Transcripts representing five genes known to be associated with pigmentation were reliably aligned against large mink genomic contigs derived from BAC clones. Sequence comparison between transcript and genomic data revealed seven SNPs. In this study, we generated and annotated mink transcript sequences representing more than 16 000 known genes. This is the first comprehensive transcriptome for the American mink genome, which will facilitate further development in mink expression profiling studies and provide a good annotation basis in the perspectives of a whole genome sequencing project. The project was deposited at EMBL database with the accession number PRJEB1260.
Keywords: de novo assembly HiSeq2000; paired end sequencing.
© 2014 Stichting International Foundation for Animal Genetics.