Select Sires

SELECTIONS - Fall 2004
Genetic Evaluations for Health and Fertility Traits - Helpful but Handle With Care

by Charles Sattler, vice president, genetic programs

Chuck Sattler

There is growing interest among dairy producers to use A.I. sires that will improve the health and fertility of their herds. This is understandable as producers are continually faced with the challenge of maintaining or reducing the cost of producing milk.

Another contributing factor is the introduction of genetic evaluations by USDA for several of these traits. Rankings of dairy sires for these traits are encouragement for us all to figure out the appropriate role of this information in our selection decisions.

How to use these new traits in breeding programs has been a topic of debate in the genetics community for a number of years. These values have gradually become more accepted and have received increasing emphasis in the TPI and Net Merit indexes. We have learned a few things and made some adjustments. Now that we have a reasonable track record with these traits, it’s worthwhile to look back and see how well these evaluations have performed.

Somatic Cell Score

USDA has provided somatic cell score (SCS) evaluations since 1994. The SCS data comes from monthly DHIA test results. USDA compiles this data and uses virtually the same animal model evaluation procedures as used for the production traits to calculate SCS genetic evaluations. SCS is a useful trait because it is an indicator of clinical and sub-clinical mastitis. The correlation between SCS and the incidence of mastitis is 70 percent.

Predicted transmitting abilities (PTA) for SCS generally range from +2.60 to +3.70. Higher values indicate bulls with daughters that tend to have higher somatic cell counts therefore bulls with lower SCS PTAs are preferred. SCS is a trait that is influenced more by management and less by genetics. For this reason, SCS PTAs tend to have lower reliabilities than the customary production and type evaluations. New graduates will enter the lineup with reliabilities for SCS PTA at about 70 percent and these will remain pretty steady until the second-crop daughters arrive.

Productive Life

Productive life (PL) evaluations are provided by USDA and also have been available since 1994. PL evaluations rank sires based on the longevity of their daughters and use the lactation records gathered by DHIA to indicate how long cows remain in the herd. Each cow gets credited with the number of months she is lactating up to when she reaches 84 months old. Cows receive a maximum of 10 months of PL per lactation and the dry period is not counted.

It takes a long time for cows to express actual productive life so PL evaluations allow information from a bull’s production proof, type proof and others to influence his PL proof. This procedure allows the PL evaluations to reach higher reliabilities earlier. Early in their careers, A.I. sires will have PL evaluations based solely on pedigree data and data from these indicator traits. Then, as the daughter PL data accumulates, their proofs become more heavily influenced by the actual longevity of their daughters.

PL PTAs generally range from -4.0 to +4.0 and represent the number of months of additional lifetime that can be expected from a bull’s daughters. PL is influenced more by management than by genetics. It also is observed only once during an animal’s lifetime so there are fewer data points to use to predict an animal’s genetic merit. Reliability for PTA PL will start around 55 percent on new graduates and will gradually increase as productive life data comes in.

Daughter Pregnancy Rate

Daughter Pregnancy Rate (DPR) PTAs are fairly new. USDA released the first DPR evaluations in February 2003. The DPR evaluations use freshening dates to calculate the number of days open in each lactation and then rank bulls based on how quickly their daughters breed back. By looking at days open, the DPR PTAs take into account how quickly after calving cows come back into heat and also their conception rate when bred.

DPR values range from +3.0 to -3.0 with higher values preferred. They represent the relative percentage of eligible cows that become pregnant during each 21-day cycle. Each point of DPR corresponds to four days open.

The genetic control of fertility is notoriously low but with DPR we are able to gather several observations for each cow during her lifetime. Because there are more observations of DPR than for PL, the evaluations reach a similar level of reliability in a bull’s first-crop proof. Holstein sires will enter the marketing lineup with DPR reliability of 50 to 55 percent and then gradually climb to about 60 percent reliability over the next several proof runs.

Learning From The Past

One thing health and fertility traits have in common is that the genetic evaluations have low reliabilities until bulls reach their second-crop proofs. How much fluctuation is there in these low reliability proofs? Table 1 provides a summary of proof changes since February 2001 (February 2003 for DPR evaluations) to provide an indication of how much change you might expect as these PTAs reach high reliability level. The bulls summarized in this table were those that were below 90 percent reliability in February 2001 and were above 95 percent reliability in August 2004.

	PL	SCS	DPR	Milk
Correlation between first- and second-crop proof	63	81	77	85
Average first-crop reliability	66	71	75	87
Average change in PTA	-0.9	+0.03	+0.3	-70
Range of changes	-3.7 to +2.9	-0.32 to -0.28	-2.3 to +2.4	-925 to +705
Number of bulls in comparison	69	137	47	262

A helpful statistic in analyzing these results is the correlation between the first-crop and second-crop PTA. This correlation indicates how well the early PTAs predict the eventual ranking of sires when they reach high reliability levels. In theory, these correlations should be very consistent with the average reliability values of these evaluations but sometimes the actual results are different than theory predicts.

It’s interesting to note that early SCS evaluations turn out to be very good predictors of future results. The average change for PTAs is very close to zero and the correlation between early and later proofs is higher than might be expected from published reliabilities.

The early DPR evaluations also appear to be fairly consistent over time. DPR evaluations have not been around very long so we’ve only been able to track 47 bulls. To this point, it appears the DPR evaluations are working fairly well.

PL evaluations have not been very consistent over time. Some of this may be expected because PL values do have lower published reliabilities. But, the correlation between the early and later proofs is considerably lower than the other traits and shows they are less useful at predicting the true genetic merit of bulls.

Helpful But Use With Caution

Genetic evaluations for health and fertility traits have lower reliability levels than production traits. Because of this we should expect more proof fluctuation and re-ranking of sires in these areas. These results show the proof changes that can occur in these traits and these changes can be substantial. It also shows that SCS and DPR evaluations are more useful than PL evaluations in identifying true genetic differences between bulls in the early stages of their careers.

Some producers like sires to be above set limits before using them. Caution needs to be used when establishing these levels for health and fertility traits. As limits are added for more traits this approach naturally becomes very restrictive. Also, with more proof fluctuations in health and fertility traits, these limits are not quite as effective at focusing in on the right bulls. If limits are used, set these at levels below breed average. This way you will exclude the very worst bulls in these areas but still allow the opportunity to use bulls that may be slightly below average in one area but offer extreme improvement in another.

Using a selection index is still the best strategy for making use of health and fertility traits in your breeding program. Figuring out how much emphasis to place on each trait is complicated. These traits are included in Net Merit, Cheese Merit and Fluid Merit as published by USDA. The Holstein Association TPI and Jersey Association JPI use these traits as well. All of these have been constructed with a lot of thought and are based on expert advice. These results suggest we may want to consider placing more emphasis on the individual traits that lead to improved longevity where the information is available early and where the evaluations are performing well (production, SCS, DPR) and less emphasis on the overall longevity trait PL.

Return to SELECTIONS Table of Contents

Select Sires Inc., 11740 U.S. 42 North, Plain City, Ohio 43064 / Phone: (614) 873-4683 Fax: (614) 873-5751