Beef Carcass Genetics
Dr. Art Linton
Extension Beef Scientist
What Traits to Select?
Genetic change of beef cattle is a relatively slow process at best. With the
generation interval of cattle at about five years, it is critical that we design our
selection program carefully. This means first giving serious consideration to
which traits to select for and then judiciously choosing the cattle that fit our
selection criteria. Today we will focus primarily on our selection objectives.
Research tells us that the more traits we select for the slower will be the
rate of progress for each. With this in mind the question each of us needs to ask
first is: “Which traits should I select for?” Factors determining the answer are:
1. What is the heritability of the traits we may be selecting?
2. Which traits have the greatest influence on the profitability of our
3. What is the relationship of each trait to other economically important
4. How easily and how expensively can the trait be measured?
5. How much variation is there in the trait among animals?
For our discussion here tonight, we will focus primarily on the carcass
traits. Remember, if we select for the carcass traits it will impact the rate of
progress in other traits because of reduced selection pressure for them and
because of (positive or negative) correlated response. If time permits, we will
look at this second factor in more detail.
Table 1. Heritability (h2) Of Beef Cattle Carcass Traits
Trait Number of studiesa Weighted Mean h2 b
Backfat 26 .44
Ribeye Area 16 .42
Slaughter Weight 52 .41
Carcass Weight 19 .23
Dressing Percentage 13 .39
Cutability 12 .47
Lean:Bone Ratio 4 .63
Marbling Score 12 .38
Warner-Bratzler Shear Force 12 .29
Sensory Panel Tenderness 3 .13
(Adapted from Koots et al., 1994a and Green, 1999)
Number of research studies represented.
Average heritability of trait, weighted by number of observations in studies.
In general, carcass traits all tend to be relatively high (above .40) in
heritability. Therefore, if we can measure these traits and there is adequate
variation within each, then we can expect to make reasonably rapid progress for
them. We also know that with grid-based marketing both yield and quality grades
impact carcass value. Yield grade is cutability and marbling is essentially the sole
determinant of quality grade. Large premiums for Choice quality grade
carcasses versus Select carcasses and substantial discounts for Yield Grade 4
carcasses provide incentive for beef cattle breeders to select for carcass traits.
Marbling may only account for a small share of the variation in palatability of
cooked beef products and is less important than tenderness, but it serves as an
“insurance policy” for eating satisfaction and is much more easily measured.
Typical grid pricing assumptions may be as follows with Choice, Yield
Grade 3 as our base:
Prime premium (above Choice) $6.00cwt.
CAB (above Choice) $3.00
Choice-Select spread $10.00
Standard discount $-15.00
YG 1 premium $3.00
YG 2 premium $1.50
YG 3 base $0.00
YG 4 & 5 discount $-25.00
Put these on carcasses weighing 800 pounds and they add up in a hurry.
Historically, carcass data was very difficult and expensive to obtain. It also
significantly increased the generation interval because it required progeny
testing, so it reduced dramatically the annual genetic progress. This has changed
dramatically with the refinement and greater availability of ultrasound technology.
It is now possible to assess carcass merit differences with a high degree of
accuracy at a young age and at reasonable cost.
Let’s take a look at how this information is being used by Angus breeders
nation-wide. The Angus breed has done a remarkable job of collecting useful
carcass information, first through a structured progeny testing program and more
recently through the collection of ultrasound data. The number of ultrasound
records in the Spring 2005 Angus Sire Evaluation totals 400,865 cattle as
compared to 78,532 cattle with actual slaughter carcass data. Figures 1 and 2
give the genetic trend for percent ultrasound intramuscular fat (which is basically
equivalent to marbling score and is abbreviated as %IMF) and for ribeye area
(REA) by comparing EPDs for those traits with the sire birth year. Both trends are
positive and, just as importantly, the rate of progress appears to be increasing. If
we were to look at annual registrations by sire, I am sure we would find that the
most heavily used bulls in that breed have superior carcass EPDs. This is a
good example of information being available and being used by Angus breeders
to dramatically improve their product. Obviously, one of the factors driving this
tend is the price incentive for cattle meeting the certification standards for
Certified Angus Beef (CAB) status.
To help put this information in perspective, the Spring 2005 breed average IMF%
EPD is .01% and the standard deviation is .14%. Over the last 13 years included
in this table the merit of the sires represented has increased by about two-thirds
of a standard deviation.
The 2005 breed average REA EPD is .03 sq. in. and the standard
deviation is .20. Sires born in 2003 are almost .15 higher than those born in
1980. The most impressive thing to note is that almost all this progress has
occurred in the last 10 years.
It is also worthy to note that significant progress has been made in both
traits in spite of the negative genetic correlation between them. Statistical
analysis of this ultrasound carcass data yielded the following results.
Table 2.Heritabilities and Genetic Correlations Based on Angus Ultrasound Data
Trait Wt. IMF REA Fat %RP
Weight (Wt.) .30 -.02 .52 .13 -.14
%IMF (IMF) .31 -.04 .22 -.16
Ribeye (REA) .38 .18 .59
Fat Thickness (Fat) .39 -.49
% Retail Product (%RP) .39
*400,865 Angus Records, Spring 2005
Heritabilities are given in bold of the diagonal
Genetic correlations between traits are off diagonal.
This information tells us that if we select just for marbling score, we can expect
a. decrease ribeye area very slightly (-.04),
b. increase subcutaneous fat (.22),
c. decrease retail product (-.16).
Likewise, if we select just for ribeye area, it will:
a. increase the percent of retail product (.59),
b. increase fat thickness (.18).
We will need to select both traits if we want to improve both traits (i.e. improve
net carcass merit). Fortunately, even though the genetic correlation between the
traits is negative (-.04), it is low enough to allow simultaneous change in the
desired direction for both traits. Just a word of caution, however: summaries of
research data have produced higher genetic correlations than analysis of breed
carcass data. Koots et al. (1994) reported a negative genetic correlation between
marbling score and ribeye area of -.21. Such a stronger relationship makes
simultaneous improvement in both traits more challenging, but still not
The current beef quality grades do not effectively segregate steaks by
levels of tenderness, although CAB appears to be more tender on average than
USDA Choice and Select. Tenderness is measured two ways in research trials:
objectively with the use of a Warner-Bratzler Shear Force (WBSF) instrument, or
subjectively with the use of a trained taste panel. Marbling has a genetic
correlation with WBSF tenderness of -.56. In other words, marbling accounts for
about 30% (R2) of the variation in tenderness as determined by WBSF. The
many attempts to develop a non-invasive technique for measuring genetic
differences in carcasses to date have not proven successful. That explains why
we still rely on marbling to evaluate carcass quality, even though it has serious
Recently an extremely large and comprehensive Carcass Merit Project with
multiple objectives and several research institutions cooperating was conducted
with funding using beef checkoff dollars through the Cattlemen’s Beef Board. A
couple of the research objectives related specifically to tenderness:
1. Develop procedures for collection of information necessary to develop
EPDs for carcass merit traits, particularly tenderness EPDs.
4. Validate DNA markers to be used in marker-assisted selection programs
for improvement of carcass merit traits.
This huge project included 8,500 progeny of the most widely used sires
within 14 breeds. Researchers used DNA markers to validate the association of
QTLs (quantitative trait loci) with WB shear force measures, sensory panel traits,
and standard carcass traits. The project was not intended to provide, nor did it
allow for across-breed comparison. Scientists from the U.S.D.A., Texas A & M,
Kansas State University, Colorado State University, and Cornell University were
involved in the project.
Two of the markers evaluated had a significant influence on WBSF
measures, indicating that these markers can be used to help improve overall
product tenderness. However, the “best” one of these markers, QTL 6, only
accounted for 12%of the phenotypic variation in WBSF. The other marker with
value accounted for 6% of the WBSF phenotypic variation. Remember that this
was with an extremely diverse (14 breed) genetic population. So, while this may
be a step forward, it is a small step at best. Currently available DNA tests for the
“marbling gene” and the “tenderness gene” only account for a small proportion
(perhaps10-15%) of the genetic variation.
Ultrasound-based marbling EPDs are more valuable than currently
available DNA markers for documenting genetic differences in marbling. In the
absence of another practical measure of tenderness, the DNA tenderness
markers remain the best tool for genetic improvement of tenderness. Look for
marker-assisted EPDs for tenderness to be developed in the near future using
ultrasound IMF EPDs and DNA markers. Our knowledge of breed differences in
tenderness can also be useful.
The Simmental breed association has made a real effort to identify sires
with tenderness differences. There are 102 Simmental bulls with progeny
tenderness data. The average Simmental bull has a .09 marbling EPD. The top
10 tenderness bulls among the 102 with tenderness data have a marbling EPD of
.16. The toughest 10 bulls have a marbling EPD of only .03. As you can see,
marbling EPD is a crude predictor of tenderness, but is certainly not a guarantee.
Relationship of Carcass Traits and Other Important Traits
Genetic correlations occur when genes tend to influence more than one
trait. These correlations may be either desirable or undesirable. It is important to
know both the direction and the strength (size) of the correlations. They tell us
what will happen to other traits when we apply selection pressure. An example of
a beneficial genetic correlation is growth rate and feed conversion. Feed
conversion is costly and difficult to measure, but the beneficial correlation with
growth rate means that selection for growth rate will also result in improvement in
Some carcass traits are negatively correlated to each other (marbling and
leanness) and to other traits of economic importance. These antagonistic genetic
correlations make progress from selection more difficult and the response to
selection smaller and slower to achieve. It is important to be aware of these
negative relationships when designing a selection program so one recognizes
what the challenges and limitations are.
Table 3. Antagonistic Genetic Correlations Among Traits
Traits Genetic Correlations
Calving Ease/Birth Weight -.74
Marbling/Yearling Weight -.33
Marbling/Shear Force -.31
Important antagonistic genetic correlations also exist between
reproduction and some product traits. Some of these relationships should be
fairly obvious when we consider breed differences for carcass and reproductive
traits. It is important to remember that these relationships exist and may present
challenges as we attempt to improve overall herd profitability.
Table 4. Genetic Correlations Between Reproduction and Carcass Traits
Female Traits Fat Trim Wt. Retail Product Wt.
Age at Puberty -.29 .30
Conception/Service .21 .28
Calving Difficulty -.31 -.02
Birth Weight -.07 .30
Mature Weight -.09 .25
Selection for Antagonistic Traits
National sire evaluation programs have been a great tool for beef
improvement. They have enabled rapid progress in single traits. But, even more
important, they have helped identify those outlier bulls that excel in antagonistic
traits. The classic examples are those bulls that combine calving ease (or low
birth weights) and rapid post-natal growth. Dr. Jim Gosey, University of
Nebraska, performed an interesting sire sort of bulls in the Fall 2003 Angus Sire
Summary with ultrasound carcass data to illustrate just how challenging it
may be to make progress for antagonistic traits. First, he compared bulls that
were extremely high (three standard deviations above breed average) for either
marbling or ribeye area.
Table 5. Carcass EPDs of Selected Sires with Ultrasound Data, Fall 2003a
Group Marbling REA Fat % RP YW Milk
Angus Average .05 .07 .00 -.01 65 17
Marbling (37) .50 .14 .02 -.04 72 21
REA (32)bc .08 .74 .00 .58 88 20
4,051 Total sires represented
Mean plus 3 standard deviations
Number of sires in parenthesis
If one examines the extreme outlier bulls for marbling they tended to be:
Above average for ribeye area, growth and milk
Fatter and below average for retail product
Extreme bulls for ribeye area tended to be:
Above average for marbling
Above average for % retail product, yearling weight and milk
Now let’s raise the bar and look at the number of bulls that are outliers for
Table 6. Number of Outlier bulls for Both Marbling and Ribeye Areaa
Trait +1 SDb +2 SDb +3 SDb
Marbling & REA 164 22 4
Marbling & REA (.50 acc) 140 22 4
4,051 Sires represented, Fall 2003 Angus Sire Summary
It quickly becomes obvious that although there may be “outlier” bulls that defy
some of the antagonisms between traits, they are very rare and, short of artificial
insemination, they don’t exist in sufficient numbers to have an immediate impact
on the commercial segment of the beef industry.
Dr. Gosey repeated this exercise for multiple traits. He asked how many
Angus bulls were simply above average for marbling, ribeye area, weaning
weight, yearling weight, scrotal circumference and below average for birth
weight? The answer was 32 or one percent. He raised the requirement to being
one standard deviation above average for all the traits, and found that reduced
the number to 10 or .3%. When he raised the requirement to two standard
deviations above average for all traits, he eliminated all the bulls.
Nobody said breeding cattle was easy. While the tools available keep
getting better, we keep asking more and more of our cattle by selecting for an
ever-increasing number of traits, several of which may be antagonistic.
One additional option for improving multiple traits simultaneously is the
use of indexes. Several of the beef breed associations have just created what
they refer to as bioeconomic indexes. These indexes attempt to describe a
number of different beef cattle production scenarios from cow-calf, to feedlot, to
harvest, and attempt to rate breeding cattle on net economic merit for each
enterprise. For example, one of the indexes recently introduced by the American
Angus Association is the Carcass Grid Value Index (represented by $G) which
combines quality grade and yield grade attributes calculated on a three-year
rolling average of industry grids. These indexes are another selection tool that
can be used for comparing individual animals. See what is available for the
breed of interest, but be sure it includes the traits of interest before using one.
Heterosis and Breed Complementarity
There is one tool remaining which I have yet to mention. That is the
utilization of multiple breeds in a specialized crossbreeding system.
Complementarity allows us to match the strengths of one breed to weaknesses
of another. The classic example of this concept is illustrated by the improvement
in net merit achieved with British X Continental crossbred steers that benefit from
the marbling input of a British breed and the lean muscle growth of a Continental
breed. Add the benefit of heterosis to the cowherd which reduces the risk that the
cows may not be adapted to the varied resource environments in which we
expect them to perform. Heterosis is also the closest we can come to a
guarantee for a high level of reproduction in our cow herd.
1. Joint improvement in marbling and lean muscle growth will be limited
by the negative genetic correlation between the two traits.
2. The number of sires with genetic estimates for carcass traits will
continue to increase due mainly to data collected via ultrasound.
3. DNA markers for major gene effects hold promise to supplement
traditional selection tools to yield more precise selection for carcass
4. Crossbreeding can be used to temper antagonistic carcass traits
through breed complementarity.
5. Heterosis is the best tool to maintain cow productive performance and
fitness to the environment while attempting to change carcass traits.
American Angus Association. 2003. Fall 2003 Sire Evaluation Report.
Angus Carcass Evaluation and Angus Body Composition Genetic
Evaluation Using Ultrasound Measures. St. Joseph, Mo. October, 2003.
Gosey, J. A., 2004. Selecting for carcass marbling and muscling – benefits and
pitfalls. Range Beef Cow Symposium. Mitchell, NE. December, 2004.
Green, R.D., T.G. Field, N.S. Hammett, B.M. Ripley and S.P. Doyle. 1999. Can
cow adaptability and carcass acceptability both be achieved? Proceedings
of Western Section of American Society of Animal Science. Provo, Utah,
Koots, K.P., J.P. Gibson. C. Smith and J. W. Wilton. 1994a. Analyses of
published genetic parameter estimates for beef production traits. 1.
Heritability. Animal Breeding Abstracts 62:309.
Koots, K.P., J.P. Gibson, C. Smith and J. W. Wilton. 1994b. Analyses of
published genetic parameter estimates for beef production traits. 2.
Phenotypic and genetic correlations. Animal Breeding Abstracts 62:825.