Canadian Dairy Network (CDN) held an Open Industry Session on Wednesday, October 24,
2018 at the St. Hyacinthe Congress Centre, which was followed, as usual, by a meeting of the
Genetic Evaluation Board (GEB) the following day. This communication specifically provides an
update on the discussions regarding the publication of Direct Genomic Values (DGV) by CDN
since it was a key topic discussed at length during the Open Industry Session. A complete
executive summary including all actions and recommendations of the GEB will be circulated in
the near future. The CDN Board of Directors will consider all such recommendations for
approval at its meeting scheduled for Monday, December 10, 2018.
- The Open Industry Session was well attended with over 80 participants, including many
breeders as well as industry personnel. CDN extends a special appreciation to those
breeders who took the time to attend this meeting and share their thoughts.
- In advance of the meeting, CDN and Holstein Canada organized meetings and discussions
with key advocates in favour of keeping the publication of Direct Genomic Values (DGV) to
gain a better understanding of the various perspectives of Canadian breeders. As a result
of these discussions, CDN conducted additional analysis to assess the potential benefits of
DGV for breeders to make selection and mating decisions, compared to using the official
Genomic Parent Averages (GPA) of young bulls and heifers.
- In brief, the CDN presentation of analysis results included the following key conclusions:
- The DGV scale for any trait, including LPI and Pro$, is wider than the scale for GPA,
which means they are not directly comparable. It also means that the most elite
animals in the breed will have DGV higher than GPA for marketing purposes.
- The group of highest genomic young bulls available in A.I. are almost identical with
similar rankings based on either DGV or GPA. The same is also true for the ranking of
heifers in herds of breeders that have been involved for decades in herdbook
registration, milk recording, type classification and the use of A.I. sires.
- In terms of predicting the future progeny proof of genomic young bulls, the scale of
GPA is more appropriate and more accurate than DGV.
- When considering high ranking genomic young bulls, the strategy of giving preference
to those bulls with the highest difference of DGV minus GPA is not effective for
identifying the most promising sires once they are progeny proven. In fact, such an
approach ends up selecting genomic young bulls that have an lower Parent Average.
- In terms of using a herd’s heifer genomic evaluations to predict future lactation and/or
classification performance as a cow, both GPA and DGV have equal levels of
accuracy, both being superior to using Parent Average (PA) alone.
- Another strategy used by some breeders when making sire selection decisions is to
identify genomic young bulls with the highest, most extreme, DGV with the goal of
producing progeny that also have extreme genomic evaluations. Such extreme young
bulls and heifers are necessary for breed improvement and provide important
opportunities for impacting rates of genetic progress. The CDN analysis conducted to
assess the benefits of this approach showed that extreme genomic young bulls based
on GPA produced a higher proportion of extreme progeny compared to results based
on extreme DGV.
- Based on the discussions and opinions expressed, there was general recognition among
those in attendance that no scientific evidence has been found to indicate that DGV offers
any benefit over GPA for making selection and mating decisions. Given this conclusion, it is
clearly understood by all industry partners, especially CDN, breed associations and A.I.
organizations, that a concerted and collaborative communication effort is required across
the country to breeders, producers and industry personnel.
- Regardless of the technical results from the CDN analysis, those in favour of maintaining the
publication of DGV stated that doing so (a) maintains the current transparency of data
available publicly; (b) attracts international breeders to the CDN web site, which helps to
promote LPI globally; and (c) leaves the choice of whether to use and how to use DGV in
the hands of each individual breeder.
After giving due consideration to all of the above and recognizing that the GEB is an advisory
committee to the CDN Board of Directors with the mandate of making science-based
recommendations, a motion was duly passed to maintain the direction of the previous GEB
recommendation to no longer publish Direct Genomic Values (DGV) and to exclude such data
from all outgoing files. GEB members consider the timing for any implementation of this action
is the responsibility of the CDN Board of Directors but it should not be any earlier than the
stated target date of April 2019.
If there are any questions, concerns or comments regarding the above recommendation of the
Genetic Evaluation Board, please feel free to contact CDN Board Chairman, Norm McNaughton
(Norm.McNaughton@gmail.com), GEB Chairman, Brian Anderson (firstname.lastname@example.org) and/or
CDN General Manager, Brian Van Doormaal (Brian@cdn.ca).