The base for U.S. genetic evaluations will be updated, effective with the April 7, 2020, triannual evaluations.

The genetic bases to which (most) dairy traits are expressed in the United States have been updated every five years since 1980. With the base change, users of genetic evaluations may become aware that the standards they set for choosing service bulls or valuing females in the past may no longer meet the genetic quality to remain competitive, due to genetic progress.

Since 1980, some have suggested that the base should be updated more often. A few have lobbied for a fixed base, or one that’s never updated. The reasoning for the latter is that if the best bulls are chosen, the magnitude of the numbers are not particularly important and all evaluations are comparable regardless of when published.

For the last base change in 2015, the average predicted transmitting abilities (PTA) of cows born in 2010 were set to zero. Progress continued to be made for most traits, as shown by Table 1 with the PTAs of cows born in 2015. These milking cows born in 2015 define the new base. With the April evaluations, their PTAs will be set back to zero. Stated differently, the averages in Table 1 will be subtracted from the current PTAs of all animals. These are the changes in PTAs expected in April.

Because gains were made across five years for most traits, most of these PTAs will be lowered by the amount shown. However, if the trends were unfavorable, the PTAs will generally increase. The exceptions can be for somatic cell score (SCS) and the four calving traits which may do the opposite because lower values are preferable for these traits. The average PTAs in the table are the differences in transmitting ability for animals over the five-year period. A note of caution, these will not be the exact changes coming because all will be recalculated before the April 2020 run using more complete and current data. Any updates in the traits’ variation will also cause these approximations to vary from the estimates presented.

Key progress points demonstrated in Table 1 include:

• Favorable gains are shown for 81 of the 102 traits (excluding conformation), while 18 were
• The most important traits (all lifetime merit indexes) showed genetic improvement for all the breeds; the largest gains were for Holsteins, Jerseys and Ayrshires. Thus, the merit indexes for all breeds will be lowered in
• Genetic gains were made in all three yield traits (milk, fat, protein) for all breeds. Gains were particularly impressive forHolsteins and Jerseys; so the base change will reduce PTA milk for these breeds by about 492 and 524 pounds,
• PTAs for fat and protein will be adjusted down by about 18 to 25
• Changes in PTAs for somatic cell score (SCS) will be small (-.01 to +.02) for all breeds except Holsteins which will increase by 0.08 due to their progress in lowering
• PTAs for productive life will be reduced by about 0.6 to 9 months for Guernseys, Holsteins, Jerseys and Milking Shorthorn due to increasing their genetic capacity for longer life.
• Unfortunately, 13 of the 18 fertility estimates showed unfavorable changes over the five years; only Holsteins improved for all three
• PTAs for cow livability, launched in August 2016, improved for three of the six breeds (0.74 for Holsteins).
• Resistance against diseases in Holsteins improved for five of the six
• PTAs increased for 80 of the 90 breed conformation traits, which indicates that selection has been for the higher scores. In most cases this probably was desirable, but in others, perhaps not. The 10 traits with PTAs that did not increase were ones that had an intermediate

Table 1. Difference in predicted transmitting abilities (PTAs) of cows born in 2015 compared to those born in 2010. PTAs will decrease by these amounts to implement the 2020 genetic base change1.

Revisions made 2.21.2020 affect the genetic base numbers for Holstein conformation traits, reflecting revised Holstein Association USA requirements to determine which cows born in 2015 met criteria for inclusion.

1 These estimates are traditional genetic predictions prior to modification with genomic data. Positive PTA will be lowered by this amount to satisfy the new base. Conversely, a negative value means the PTA will be raised by this amount. Estimates can change as additional records are added and trait variation is included. The red color for a number indicates there was a loss for that trait between 2010 and 2015. The “ ̶ ” means a genetic evaluation is not calculated for the trait in this breed.

2Completed information for this trait is not available yet for 2015, so it is replaced by the most recent annual average available.

P erformance Differences Attributed to Genetic and Environmental Changes

The PTAs in Table 1 represent only half the genetic change achieved, as each animal only transmits half of their genes to their offspring. Table 2 shows the total changes in performance between cows born in 2015 and 2010 and an indication of how much of the changes were attributed to genetics and environment.

Key points:

• For the milk traits, all breeds but Guernsey revealed a positive contribution from both
• Genetic gain for somatic cell score (SCS) was made for three breeds, and Holsteins improved by 17.

Two breeds were unchanged, and Milking Shorthorn increased by 0.05.

• Environmental trends for SCS and productive life (PL) generally were
• Environmental trends for daughter pregnancy rate (DPR) and cow conception rate (CCR) were
• For heifer conception rate (HCR), environmental trends were negative for Holsteins, Jerseys and Milking Shorthorn.
• All six breeds showed reduced age at first calving (AFC), particularly via the environment (10 to 26 days).

Genetics reduced age at first calving up to three days.

• Phenotypic reductions in gestation length for Holsteins and Jerseys seemed surprising, especially since the genetic component for Holsteins decreased while Jerseys
• Phenotypic changes for resistance to the six health disorders introduced in April 2018 were all

Table 2. Differences in actual (phenotypic) performance between cows born in 2015 and those born in 2010 attributed to genetic (BV=breeding value1) and environmental changes.

 1 These changes are based on traditional genetic predictions prior to modification with genomic data. An “ ̶ ” means a genetic evaluation is not provided for the trait in that breed. A red number indicates there was no phenotypic improvement between 2010 and 2015, but instead apparent deterioration for the trait.

2Completed information for 2015 not available yet, so it is replaced with the most recent annual average available.

Impact of Genomics

The genomic revolution initiated in 2008 brought an increase in the rate of genetic improvement, primarily due to a reduction in the generation interval. A small portion of genomic benefits would have been revealed in the previous base change for cows born in 2010, but the current update will reflect all benefits from genomics a chieved from 2010 to 2015.

Key points:

• For illustration, 150% would indicate 50% more gain was made than in the previous five-year
• Use of genomics is responsible for the accelerated gains for milk traits shown in Table 3 for Brown Swiss, Holsteins and Jerseys, but genomics were not available for Guernseys until 2016. Ayrshires and Milking Shorthorn – having limited use of genomics – show smaller gains in milk traits than during the previous five-year
• The benefits of genomics for productive life (PL) were impressive for Holsteins and
• The Guernsey, Holstein and Jersey breeds showed larger gains (43 to 100% more) in the lifetime merit indexes for this base update, than they did during the previous

Table 3. Relative size in percentage of the 2020 genetic base changes1 compared to the base changes five years earlier (2015).

1 These approximations are based on the traditional genetic predictions prior to modification with genomic data. Cells showing an A indicate there was no gain between 2005 and 2010 (instead a loss) so the ratio of gain is undefined. Cells with a B indicate there was no gain between 2010 and 2015. The ( – ) means that a genetic evaluation is not provided for the trait in that breed.

2Completed information for 2015 not available yet, so replaced with the most recent annual average available.

Percentage of Change Attributed to Genetics

To answer the question of what is contributing to the phenotypic improvement being made in dairy production traits, Table 4 was derived from the information in Table 2 for traits that have had evaluations initiated since 2008. The genetic contribution averaged 45% but was greater (averaged 69%) for the three traits with the greatest emphasis in net merit index (NM$) and for Holsteins (71%), the breed with the largest population.

Table 4. Percentage of the change in phenotype attributed to genetics for cows born in 2015 compared to those born in 2010 for traits with published evaluations initiated before 20101.

1 These changes are based on the traditional genetic predictions prior to modification with genomic data. A “0” indicates there was no genetic improvement in five years. A “100” in red indicates there was genetic improvement and the phenotypic change was unfavorable. A “100” in black indicates that the genetic improvement exceeded the phenotypic change. See https://aipl.arsusda.gov/reference/base2014.htm for the previous base change report from December 2014.

When the base is changed every five years, most PTAs are lowered – and the standard deviations (SD) are also updated. In mostcases, the variation increases. Yield and SCS records are adjusted for variance within herd and year to keep the same SD as the base year using SD ratios shown in Table 5.

Table 5. Ratio of trait SD for base cows born in 2015 vs. those in 2010. The PTAs will be expanded (or contracted) by these ratios.

Table 6. Ratio of trait SD for base cows born in 2015 compared to Holsteins. The PTAs will be expanded (or contracted) by these ratios.

The Bottom Line

Advocates for improving sustainability and eliminating world hunger should be amazed to see the changes in productivity in U.S. dairy.

Greenhouse gases are being reduced per unit of product because of greater production per animal. We are seeing significant changes in the animals’ appearance and health as well.

Since we’re approaching another base change, this may be a good time to remind dairy producers to adopt genetic selection strategies that could virtually eliminate any complacency of decisions between base changes. For example, if selection is based on standards like percentiles (recalculated every run) or by simply selecting the top-ranked bulls on an economic index of their choice, forward progress would occur, devoid of any delays.