Toward the Construction of an Accurate Reference Sequence for the Domestic Cow (Bos taurus) Genome Using Single Molecule Analysis

Cattle have played an important role in human history by providing draft power, sustenence for human consumption.  Cattle, as a model organism, has greatly contributed to our knowledge of mammal genetics, biochemistry, and general biology, especially for ruminants. The cattle genome has been sequenced and its initial sequence assembly was published in 2009 by the Bovine Genome Sequencing and Analysis Consortium (BGSAC). This work utilized Sanger sequencing reads (~7.1-fold) coverage realized through a combination of bacterial artificial chromosome (BAC), whole genome shot-gun (WPS) sequence data and the integrated bovine genome map. However, an alternative bovine genome sequence assembly (UMD) was generated by the Center of Bioinformatics and Computational Biology (CBCB), University of Maryland, using the same sequence reads and map data, but a different assembly strategy that utilized the paired-end BAC sequence information, and synteny to human genome. Comparison between these two bovine genome sequence assemblies showed substantial differences in assembly errors, genome segment inversions, chromosomal placements and gaps due to the use of different assembly algorithms, and the insufficient marker density in the cattle integrated genome maps. The differences presented by these genome sequence assemblies have already affected comparative genomic studies among cattle breeds or populations because some structural variants are associated with phenotypic traits and diseases. Consequently, when these assemblies were used as reference genome sequences for comparative genomic studies, additional work was required for disambiguation of structure variants within those dissimilar regions harbored by the two assemblies. Therefore, another independent whole genome high resolution physical scaffold would provide a much needed genome resource to resolve the differences between the two bovine genome sequence assemblies, place and orient orphan sequence contigs or scaffolds, correct the misassemblies, characterize the sequence gaps, and facilitate the construction of an accurate bovine genome reference sequence for future research. Accordingly, we have constructed a whole genome high-resolution optical map for the sequenced Dominette L1 cow genome to guide the many stages of genome assembly and validation. This optical map has offered a uniquely effective means for resolving the differences between the two bovine genome sequence assemblies and has also recruited orphan sequence contigs to fill in gaps. Collectively, these efforts accelerate the construction of an accurate, universal reference sequence for the bovine genome.