The immediate impact of the human genome sequence on human health and wellness has been rather underwhelming. Yes, sequence information from human and/or microbial genomes is immensely useful for understanding biology. But translating that into novel medicines and diagnostics is both complex and time consuming. Meanwhile, less anthropocentric genomic studies are forging ahead, with very little hype or fanfare. These livestock genomes are providing not only hugely valuable biological information but also immediate benefits to the way livestock breeders go about their business.

The latest milestone in animal genomics is the assembled sequence of Bos taurus, domestic cattle (Science 324, 522–528, 2009;, and an analysis of >37,000 single nucleotide polymorphisms (SNPs) in 497 cattle from 19 geographically and biologically diverse breeds (Science 324, 528–532, 2009). The bovine genome follows the chicken sequence (Nature 432, 695–716, 2004) and precedes those of the pig and sheep, which are slated for release later this year and sometime in the next couple of years, respectively. All of these livestock sequences provide insights into gene function, evolution and the origins of different breeds. But they also provide something that simply cannot be exploited in humans.

Where human genomics does (somewhat rarely) shed light on human characteristics, the genetic information can be used only indirectly. It is considered immoral to breed out 'undesirable' traits in humans. Consequently, pinpointing a disease-associated allele in a human may improve our understanding of the condition and suggest a drug target or a plausible diagnostic agent, but its utility only emerges when (or if) the product development processes successfully run their course.

The case could not be more different for agriculturally important animals. For all types of livestock, genomic information is directly aligned with the means for improvement. First identify the gene(s); then breed (or splice) them in. In fact, it is not even necessary to identify the gene(s) or to understand the molecular processes behind milk yield or meat quality or the fecundity of sheep. All that is really needed is to be able to correlate the desired quality with the genetic variant and then to set up an appropriate breeding program.

Thus, working from the draft bovine sequence assembly, researchers led by Curt Van Tassell of the Agricultural Research Service at the US Department of Agriculture (USDA) teamed up with Illumina to create a BeadChip containing >50,000 bovine SNPs. This was commercially launched in January 2008, and by July last year, the dairy industry was using it to direct its breeding approach. Similarly, with <50% of the sheep sequence available, an Illumina Ovine SNP50 BeadChip released in January is already being used to analyze a population of 20,000 sheep in New Zealand and elsewhere. Part of the reason that SNP discovery can proceed so rapidly is the increasing use of reduced representation libraries and next-generation sequencing, which provide not only SNP positions but also concurrent estimates of minor allele frequencies (Nat. Methods 5, 247–252, 2008), all at a cost of about $0.50 a SNP.

Another key difference from the human situation is that animal genomic information is not data in a vacuum. There are large databases on breeding and progeny stock for several of the main types of livestock animal, data that can be correlated with genetic data. Thus, for most animal genomes, a much bigger piece of the data jigsaw is already in place when genome sequence becomes available.

For example, New Zealand's sheep database was started in 1968 and now contains data for over 6.5 million animals on key characteristics, such as lamb survival, the number of lambs born, their birth weight, growth rates, disease resistance and meat yield and quality. For dairy cattle, the USDA has data from breed registry societies going back to 1960—over 60 million milk records covering more than ten generations of animals. The pork, beef and chicken records are less comprehensive, but even so the cataloging can go back at least as far as the great-grandparents of the current generation.

Centuries of directed improvement means that domestic animal variation is narrow, even between countries or climates. This means a genomic study of one cow or chicken is often directly relevant to other domesticated strains. Furthermore, the inbred nature of food animals makes the application of trait-associated markers more straightforward, especially when one breed predominates. This is one reason why the uptake of marker-assisted breeding has been relatively rapid in the US dairy industry, where >90% of cows are Holsteins.

The fact that at least three companies now market gene tests to breeders for economically important traits suggests demand for marker-assisted breeding exists. In March 2008, Pfizer Animal Health launched a new animal genetics division to market gene tests for quality grade, tenderness and feed efficiency. Elsewhere, Merial (Igenity) and Cargill (Metamorphix) have also put gene tests on the market for similar traits. Pricing competitively may be key to success in marker-assisted breeding and the market dynamics will almost certainly vary depending on the application of the approach. A $100 genetic test that is feasible for the sire of a dairy herd (large number of offspring and continuing benefit of raised milk yields) might be prohibitive for a beef herd (fewer animals per sire and a one-off product). Concerns are also emerging over the ability of companies to acquire intellectual property to protect the genes in their tests (see p. 496).

Nevertheless, the beauty of marker-assisted breeding is that it is likely to be less disconcerting to both the public and regulators than transgenic or cloned animals. Animal genomics may just be the accelerant that animal biotech needs to grab the spotlight from its more showy but often unproductive human health counterpart. Certainly, if marker-assisted breeding lives up to its early promise, it could change the face of animal health, welfare and productivity.