A New and Improved Rainbow Trout (Oncorhynchus mykiss) Reference Genome Assembly

In an effort to improve the rainbow trout reference genome assembly, we re-sequenced the doubled-haploid Swanson line using the longest available reads from the Illumina technology; generating over 510 million paired-end shotgun reads (2x260nt), and 1 billion mate-pair reads (2x160nt) from four sequencing libraries with fragment sizes ranging from 800bp to 15Kb.  We then generated a 2.17Gb genome assembly, containing 139,726 scaffolds with N50 greater than 1.7Mb using DeNovoMAGIC, the advanced genome assembly pipeline from NRGene.  To generate a dense genetic map for each chromosome, we genotyped more than 4,000 pedigreed fish with the Affymetrix 57K SNP array and with RAD SNPs.  Using these genetic maps’ information we anchored and ordered ≈80% of the assembly scaffolds within chromosomes.  Our initial analysis indicates that the new physical genome sequence assembly and its integration with the new genetic maps results in major improvements over the current rainbow trout reference genome in all categories; including genome coverage, scaffold size, precision, and anchoring and ordering of the scaffolds onto chromosome maps.  The assembly can be further improved by merging of neighboring scaffolds using overlapping long sequences from other trout assemblies or long reads sequencing technologies.