CGs to EPDs

1
CGs to EPDs

• 2006 BIF SymposiumSponsored by Ultrasound
  Guidelines Council
• Dr. Lisa A. Kriese-AndersonAuburn University

2
Would you purchase a bull .
3
Would you purchase a bull .
This top young son of Superbull had a weaning
weight of 636 lbs
4
Would you purchase a bull .
An exceptional prospect with a scrotal
circumference of 35 cm
This top young son of Superbull had a weaning
weight of 636 lbs
5
So, why would you consider.
6
So, why would you consider
7
So, why would you consider
8
Ultrasound Basics

• Individual ultrasound measurements (adjusted) are
  as useful as individual weights and measures
• Really only mean something within the group they
  came from
• Ratios are better
• EPDs for ultrasound or carcass are best for
  selection decisions

9
Ultrasound Genetics Refresher

• Ultrasound measurements should be taken the same
  time as yearling weights
• Generally 320/330 to 410/430 days of age
• BIF says 335 to 395 days
• Entire contemporary group should be measured by a
  certified ultrasound technician
• Genetically, ultrasound measurements for carcass
  traits are highly correlated to actual carcass
  traits
• But they are not the same traits

10
Ultrasound Genetics Refresher
Genetic Correlation Estimates Among Yearling Ultrasound and Carcass Traits Genetic Correlation Estimates Among Yearling Ultrasound and Carcass Traits
UBF and BF 0.75 to 0.80
UREA and REA 0.70 to 0.75
IMF and Marbling 0.60 to 0.68
11
Ultrasound Genetics Refresher
Genetic Correlation Estimates Among Yearling Ultrasound and Carcass Traits Genetic Correlation Estimates Among Yearling Ultrasound and Carcass Traits
UBF and BF 0.75 to 0.80
UREA and REA 0.70 to 0.75
IMF and Marbling 0.60 to 0.68
Reminder if you do not report any carcass data,
individual animal accuracy can only be a high as
the genetic correlation
12
Genetic EvaluationBegins withP G E
13
In Genetic EvaluationE Contemporary Group
14
Contemporary Group

• A contemporary group is a group of animals with
  the same
• Herd
• Sex
• Birth Season
• Weigh date
• Management
• The largest contemporary group is at birth. All
  subsequent traits are subsets of the birth
  contemporary group.

15
In Simplest Terms

• BV h2 x SD
• SD Selection DifferentialSD (individual
  avg of group)

16
Example Data
No Sire Dam CG WW PW REA BF IMF
6001 1 100 1 781 486 11.5 0.30 4.25
6002 1 101 1 626 522 11.5 0.21 4.21
6003 2 102 1 845 357 11.3 0.28 4.35
6004 2 103 1 713 482 11.2 0.18 2.89
6005 2 104 2 663 397 10.1 0.17 4.59
6006 3 105 2 720 532 10.6 0.44 5.20
6007 3 106 3 720 539 12.5 0.19 3.82
6008 4 107 2 841 562 15.1 0.32 3.90
6009 4 108 3 865 510 11.9 0.31 4.44
17
Data Analysis

• Multiple-trait animal model
• WW, PWG, UREA, UBF, IMF in analysis
• Examine Sire Rankings for above traits

18
Sire EPD Rankings
Trait Trait Trait Trait Trait Trait
WW Milk PWG UREA UBF IMF
4 4 4 4 2 1
2 2 1 1 1 3
1 1 3 3 4 2
3 3 2 2 3 4
19
Example Data Just 1 CG
No Sire Dam CG WW PW REA BF IMF
6001 1 100 1 781 486 11.5 0.30 4.25
6002 1 101 1 626 522 11.5 0.21 4.21
6003 2 102 1 845 357 11.3 0.28 4.35
6004 2 103 1 713 482 11.2 0.18 2.89
6005 2 104 1 663 397 10.1 0.17 4.59
6006 3 105 1 720 532 10.6 0.44 5.20
6007 3 106 1 720 539 12.5 0.19 3.82
6008 4 107 1 841 562 15.1 0.32 3.90
6009 4 108 1 865 510 11.9 0.31 4.44
20
With Wrong CG Definitions

• Do not have correct comparisons in data
• Sires are compared in head to head competition
  that were not
• Basics of EPD analysis is to find the group
  average and subtract the individual measurement
  from it
• Group mean is wrong!

21
Comparing Sire Ranks Wrong CG
Trait Trait Trait Trait Trait Trait
WW Milk PWG UREA UBF IMF
4 4 4 4 2 1/3
2/1 2 1/3 1/2 1 3/1
1/3 1/3 3/1 3/1 4/3 2/4
3/2 3/1 2 2/3 3/4 4/2
22
Not Reporting All Data
No Sire Dam CG WW PW REA BF IMF
6001 1 100 1 781 486 11.5 0.30 4.25
6002 1 101 1 0 0 0 0 0
6003 2 102 1 845 357 11.3 0.28 4.35
6004 2 103 1 713 482 11.2 0.18 2.89
6005 2 104 2 663 397 10.1 0.17 4.59
6006 3 105 2 720 532 10.6 0.44 5.20
6007 3 106 3 720 539 12.5 0.19 3.82
6008 4 107 2 841 562 15.1 0.32 3.90
6009 4 108 3 865 510 11.9 0.31 4.44
23
By Not Reporting All Data

• If not sending in bottom-performing cattle,
  penalize the rest
• Group means are incorrect once again
• Competition is not reported accurately

24
By Not Reporting All Data
Trait Mean All data Mean-Select Data
WW 753 85 lbs 769 75 lbs
PWG 487 68 lbs 483 72 lbs
UREA 11.7 1.4 sq in 11.8 1.5 sq in
UBF 0.27 .09 in 0.27 .09 in
IMF 4.2 .63 4.2 .67
25
Sire EPD Ranks Incomplete Data
Trait Trait Trait Trait Trait Trait
WW Milk PWG UREA UBF IMF
4 4 4 4 2 1/3
2/1 2/1 1 1 1 3/1
1/3 1/2 3 3 4 2
3/2 3 2 2 3 4
26
Not Reporting All Data
No Sire Dam CG WW PW REA BF IMF
6001 1 100 1 781 486 11.5 0.30 4.25
6002 1 101 1 0 0 0 0 0
6003 2 102 1 845 357 11.3 0.28 4.35
6004 2 103 1 713 482 11.2 0.18 2.89
6005 2 104 2 663 0 0 0 0
6006 3 105 2 720 532 10.6 0.44 5.20
6007 3 106 3 720 539 12.5 0.19 3.82
6008 4 107 2 841 562 15.1 0.32 3.90
6009 4 108 3 865 510 11.9 0.31 4.44
27
Sire EPD Ranks Incomplete Data
Trait Trait Trait Trait Trait Trait
WW Milk PWG UREA UBF IMF
4 4 4 4 2 1/3
2/1 2/1 1 1 1/3 3/1
1/2 1/2 3 3/2 4 2
3 3 2 2/3 3/1 4
28
Final Comments

• Actual ultrasound data needs to be treated the
  same as actual data from any other trait
• Must be in a comparison mode (Ratio/EPD)
• For use within a contemporary group
• May be helpful to divide data into thirds
• Always remember there can be measurement error.
  Dont believe in absolutes

29
Field Certification
Trait SEP SER Bias Corr
Rump/Back Fat 0.1 0.1 0.1 0.85
Ribeye Area 1.2 1.2 1.2 0.80
Percent IMF 1.2 1.2 0.7 0.70
30
Final Comments

• Proper reporting of contemporary groups is
  essential
• P G E is the basis for all genetic evaluation
• Dont selectively report data
• Only penalizing the top animals
• Only hurting yourself
• Dont let a trait you may not be able to
  visualize well keep you from doing the right thing

31
Additional Items

• If a breed publishes both ultrasound and carcass
  EPD values, want ultrasound and carcass EPDs to
  be similar

Trait Ultrasound Carcass
REA 0.26 0.36
IMF 0.27 0.27
Bull 16 Bull 16 Bull 16
Trait Ultrasound Carcass
REA 0.33 -0.04
IMF 0.46 0.10
Bull 110 Bull 110 Bull 110
32
Additional Items

• If breed published just carcass and you
  collect/report no carcass info, accuracy will
  only be as high as genetic correlation
• Dont single trait select!