Multiple Trait Selection for Maternal Productivity

1
Multiple Trait Selection for Maternal Productivity

• D. H. Crews, Jr., P. B. Mwansa and R. A. Kemp
• Agriculture and Agri-Food Canada Research Centre,
  Lethbridge, AB

2
What is maternal productivity?

• Reproductive rate
• Age at puberty
• Heifer pregnancy rate, calving ease and 3 yr old
  rebreed rate
• Persistence into profitable parities
• Calf growth
• Direct genetics for growth
• Maternal genetics ? milk production
• Cow maintenance requirements
• Weight
• Intake

Low heritability, Difficult evaluation
Everyone already does this
Cant measure this
3
Data and Information

• One obstacle to new index development is
  capturing information on new traits
• We are often limited by what data is routinely
  and comfortably recorded in national databases

You want to measure what?!?
4
Multiple Trait Selection

• Measure This
• Growth
• Cow weight
• Calves Reproductive years

• Evaluate This
• Maternal Productivity

5
Trait Definitions

• Direct weaning weight
• Maternal weaning weight
• Cow weight at calf weaning
• Indicator for maintenance costs
• Stayability
• Pr(3 1) probability that a female will have
  at least three calves given she became a dam

6
Multiple Trait Selection Index Objective

• The selection objective was to increase genetic
  potential to consistently wean heavy calves while
  maintaining input costs which yielded an
  aggregate genetic value function
• MPI v1 WWT v2 MLK v3 CWT v4 STY
• with resulting economic weights
• v1 1.17
• v2 0.98
• v3 -0.14
• v4 2.39

7
MPI NCE Genetic Parameters
-0.48
8
Relative Emphasis

• The relative contribution of component traits to
  variation in the index
• Weaning weight 30
• Milk 25
• Cow weight 13
• Stayability 27

9
Making the MPI User Friendly

• The MPI can be interpreted as any other EPD
• It is a combination of EPD with relative economic
  values
• The units are differences
• Information packaging is an important part of
  making indexes useful selection tools rather than
  only marketing tools
• The MPI can be expressed with any mean and
  variance
• e.g. mean 100, SD 25

10
MPI Validation

• What differences in animals are we evaluating?
• Is the MPI the same as selection for component
  traits?
• Separation of MPI evaluation into high and low
  groups
• High gt 2 SD greater than the mean MPI (n
  17,328)
• Low gt 2 SD less than the mean MPI (n 11,496)
• Comparison of standardized component EPD between
  groups
• What component traits are most related to the MPI?

11
Component Traits in High and Low MPI Groups

12
To Summarize

• The MPI was developed with the breeding objective
  to increase genetic potential to consistently
  wean heavier calves over a sustained productive
  life while maintaining input costs
• Annual or generational genetic trend is expected
  to be positive for all component traits, however,
  cow weight increases would be minimal
• As a multiple trait index, component trait values
  vary
• Animals with different component EPD have similar
  MPI

13
Summary

• The MPI places more relative emphasis on maternal
  characteristics and stayability than on growth
• Accuracy and intensity are greatly reduced when
  data (information) is lacking on grandprogeny
• MPI selection is not equivalent to selection for
  any of the component traits alone
• Validation with maternal indexes is a challenge
• MPI validation is still largely to be conducted
• The MPI needs to be expanded to include more
  components, which will come with data

14
Thank you

• dcrews_at_agr.gc.ca
• 403-317-2288