The standard theory of evolution is one of the steady changes driven by selective pressures, punctuated by occasional periods of acceleration arising from sudden, and sometimes dramatic, shifts in circumstances. That general picture holds true also for managed evolution, which is the business of this journal and of those who work in animal breeding and genetics. Judging from the rapid progress in genomic analysis and its even more rapid deployment in practice, particularly in cattle, we have now entered a period of acceleration. Current studies (e.g. Pryce et al., 9th WCGALP, 2010) suggest that with genomic prediction, it will be possible to maintain the accuracy and intensity of selection and to halve the generation interval. With the resultant possibility of doubling the rate of genetic change, is it time to look ahead and ask some questions about the long-term goals?
A good starting point for such a reflection is an important benchmark study published in 1982 by EAAP (with the full name now being the European Federation of Animal Science). With input from over 50 animal scientists, it attempted to look back to the developments in the preceding 20 years and to look forward to the end of the century. It was able to record remarkable growth in output and efficiency in most areas of European animal agriculture. Particularly in the Western free market countries, and facilitated by the growth and consolidation of the European Union, high levels of consumption were maintained with fewer animals, fewer farmers and declining costs to consumers.
A major contributor has been steady genetic improvement in all farm animal species. This is best documented in dairy cattle, where widespread use of AI facilitated intense selection on the male side. Accuracy of selection rapidly reached high levels as new models and methods for managing field data matured. The result is that for several decades now we have had linear genetic gains of approximately 1% per annum in output per animal (e.g. http://aipl.arsusda.gov/eval/summary/trend.cfm). The general abandonment of dual purpose goals has meant that this has been paralleled by the virtual disappearance of many traditional European breeds, to the point where a single breed, Holstein of North American origin, now dominates dairy production in almost every developed country.
Within that set of interconnected Holstein populations, the intense and focused selection pressures have had some negative consequences. One has been the documented decline in major fitness traits (fertility, longevity), and another has been an accelerated increase in inbreeding. Effective population size for the US Holstein has been calculated as 39 [Weigel, J. Dairy Sci. 84(E.Suppl.): E177-E188, 2001]. Given extensive international genetic exchange and the fact that comparable effective population sizes have been calculated in a number of other countries, Holstein inbreeding rates generally are now increasing by approximately 1% per generation. The declining fitness observed in most countries can be caused by this increased inbreeding or by negative correlated responses to genetic improvement in production capacity. It could also be a consequence of the greater physiological stress of production levels now exceeding 10 000 kg per cow per year in some countries.
Will the shift to genomics make these problems better or worse? Doubling the rate of genetic change, with unchanged selection practices and objectives, would undoubtedly accelerate inbreeding and would also accentuate negative correlated responses. However, if it leads to less concentrated sire-of-sire selection, that could moderate inbreeding effects. Genomic data could also be used to directly monitor changes in genetic variation, possibly leading to new ways to control the rate of inbreeding.
In addition, genomics offers the prospect of more accurate evaluation for low heritability traits, and this, coupled with increased emphasis on fitness in selection indexes, could moderate negative impacts. Harvesting the promise of genomics therefore requires substantial redesign of breeding programmes.
This leaves open the question of whether physiological limits will ultimately prevail. These limits are more acute when feed resources are restricted or where climatic effects are severe. In an attempt to restore fitness, cross-breeding has been promoted in several countries. It has been most successful where grazing systems or climatic stress impose such nutritional and physiological constraints. Perhaps genomics, combined with advanced reproductive techniques, has something to offer here. In some tropical countries, for instance, it is clear that in cattle, F1 taurus x indicus crosses outperform both parental sources and all other combinations. Their advantage seems to depend on widespread gene complementarity, well beyond simple single locus heterosis. In such circumstances, it might be possible to produce sexed hybrid embryos cheaply enough to provide continuous F1 replacements. Perhaps genomic analysis could identify specific gene combinations that would maximise not just the dominance but the epistatic components of such hybrid advantage. In temperate countries with highly intensive production, this might also be a route to effective heifer replacement programmes.
The future will be substantially different from the past, with at least as much challenge and, we hope, as much reward.