Agriculture Reference
In-Depth Information
Accuracy of Selection
received the beneficial allele, and the other sib-
ling may have received the detrimental allele. In
this case, GS would predict that the former sibling
would have a higher BV than the latter sibling
based on their different genotypes for that gene,
while parent average would predict them to have
the same value. Therefore, a breeder using GS
could select between these full siblings while a
breeder using traditional, pedigree-based selec-
tion would not be able to tell them apart until
their phenotypes were recorded.
The development of statistical methods for accu-
rately calculating EBV has been one of the major
success stories of animal breeding. A relatively
new DNA-enabled approach to calculate EBV of
animals is genomic selection (GS), which was
proposed by Meuwissen
et al
. (2001). This con-
cept is based on the idea that sections of DNA
associated with trait variation can be tracked by
DNA markers spread throughout the genome
and that this can be used to accurately predict
the EBV of an animal at a young age. The
approach essentially involves simultaneous
selection on tens or hundreds of thousands of
markers that are distributed throughout an ani-
mal's genome based on their relative effect on
specific traits. These marker effects are estimated
using the phenotypes and genotypes of a large
number of animals from a reference population.
Using this information, an animal in another
population can have its genomic BV estimated
using only genotypic information, before pheno-
types are recorded.
This technology has already been adopted
by the dairy industry to increase the accuracy of
EBV on young bull sires prior to progeny testing,
thereby enabling the inclusion of these young
males in breeding programmes earlier in their
lives, thus decreasing the generation interval. It
is hoped that this technology will be able to con-
tribute to sustainable animal breeding by provid-
ing previously absent selection criteria for traits
of importance to sustainability goals. Many of
these traits have previously been omitted from
BO due to the expense and/or difficulty of phe-
notyping; using GS, EBV become available for a
wide variety of traits once the animal has been
genotyped. It is also thought that using GS may
help to decrease rates of inbreeding per genera-
tion, because selection using this approach
increases the emphasis on the Mendelian sam-
pling of genes an individual receives from its
parents (within family selection), as distinct
from emphasizing the parent average as in tradi-
tional BLUP, which tends to select entire families
(Daetwyler
et al
., 2007). For example, two full
siblings share the same parents, and therefore,
before being phenotyped themselves, they would
be expected to have similar genetic value. But if
one of their parents was heterozygotic for a gene
affecting a particular trait, one sibling may have
Selection Intensity
Perhaps no other technology has had as great an
impact on accelerating the rate of genetic gain
through increasing the intensity of selection
than has artificial insemination (AI). Artificial
insemination technology was introduced into
the dairy industry and commercialized in the
USA during the late 1930s to early 1940s and
achieved a domestic market of 15.5 million units
in 2002. Seventy per cent of all dairy cows in the
USA are bred using AI, as are virtually all tur-
keys and chickens. Artificial insemination allows
the extensive use of high-accuracy, genetically
superior sires and plays a major role in design of
breeding programmes and dissemination of
advanced genetics. Although AI is now used
routinely in animal breeding and human medi-
cine, it was initially viewed with scepticism.
There was a fear that AI would lead to abnor-
malities, and influential cattle breeders were
originally opposed to the concept, as they
believed it would destroy their bull market
(Foote, 2002). When independent, university
research demonstrated that the technology
could be used to provide superior bulls, control
venereal disease and produce healthy calves,
subsequent adoption was swift. To put the
impact of the genetic improvement enabled by
AI in a sustainability perspective, consider that
advances in the genetics, nutrition and manage-
ment of US dairy cows over the last century have
resulted in a greater than fourfold increase in
milk production per cow, and a threefold
improvement in production efficiency (milk out-
put per feed resource input; VandeHaar and
St-Pierre, 2006). About half of this 369%
increase in production efficiency is attributable
to genetic improvement enabled by AI. As a
