Updating the Bovine Reference Genome Assembly

The first draft of the bovine genome, derived from the Hereford cow L1 Dominette 01449, was released in 2004. While far from perfect, it provided an immediate boon to bovine genomics research. Updates and new assembly releases through the years have led to great progress from that first release, but many issues still remain with the current assemblies. Recent improvements in long read technologies and assembly methods made it clear that a new de novo reference assembly could solve many of these issues. We have generated approximately 80X PacBio and 80X Illumina coverage of Dominette. A FALCON based de novo assembly was followed by scaffolding with Dovetail Genomics Chicago library/HiRise technology, the recently developed BtOM1.0 Optical Map of Dominette, and a recombination map of 59K autosomal SNPs. The scaffolded assembly was then refined with independent de novo assemblies from CANU and MaSuRCA, yielding chromosome length scaffolds with the exception of the X chromosome. Preliminary assembly statistics include an N50 contig size of 22.6 Mb and an N50 scaffold size of 89.0 Mb representing 23- and 14-fold improvements over UMD3.1 (contig N50=0.97Mb, scaffold N50=6.4Mb). Additionally, we have sequenced full-length transcripts from 30 Dominette tissues with PacBio using the Iso-Seq method to support improved annotation. A public version of the new ARS-UCD assembly is expected to be released at the beginning of 2017.

Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.