Genetic correlations between the performance of purebred and crossbred pigs

1
Genetic correlations between the performance of
purebred and crossbred pigs

• Sansak Nakavisut
• Dr.Ron Crump
• Matias Suarez
• Dr.Hans Graser

2
Introduction

• Nucleus herds test select purebreds
• But the end products are crossbreds (multipliers
  and commercial herds)
• This assumes rg between purebreds and crossbreds
  of 1
• If not 1, testing and genetic evaluation
  procedure may need some changes

3
In general

• Estimation of rg between purebreds and crossbreds
  NOT possible or NOT reliable
• Performance test records on crossbreds not
  available
• Crossbred records may be available BUT from
  different environments (nucleus VS multiplier
  herds)

4
Thai government pig breeding farms

• Crossbreds were performance tested
• Purebreds and crossbreds were produced in the
  same conditions (environments, herds, management,
  feeding..)
• Litter records available from both crossbred and
  purebred sows
• Allows reliable estimation of rg between pure-
  and crossbreds

5
Objectives

• To estimate genetic correlations (rg) between the
  performance of purebreds and crossbreds
• To validate the conventional genetic evaluation
  procedure (whether rg 1)

6
Breeding diagram
GGP
GP
PS
7
Performance test records by breeds
8
Litter records by breeds
9
Traits to be analysed

• PRODUCTION
• TDG
• ADG
• FCR
• BF
• Body Length

• REPRODUCTION
• NPB
• NBA
• LS3W
• LWB
• LW3W
• GEST

10
Statistical model for test records
assumption
11
Statistical model for litter records
assumption
12
Fixed effect model for test records
13
Fixed effect model for litter records
The same fixed effects were fitted for all litter
traits
14
Results rg (purecross)
15
Discussion (production traits)

• rg (purecross) high enough gt testing and
  selecting of purebreds in the nucleus can
  genetically improve the production traits of
  crossbreds in multiplier and commercial herds
• To reduce cost of unnecessary test of crossbreds
• Conventional testing of purebreds is validated

16
Discussion (reproduction traits)

• rg (purecross) NOT 1 (although positive) gt
  selecting on purebred litter records can
  genetically improve the reproduction of crossbred
  sows BUT not as efficient as incorporating
  crossbred records in the genetic evaluation
  procedure.
• Crossbred records readily available in most
  multipliers with no extra cost gt combine them
  with purebred records to produce EBVs for litter
  traits

17
Conclusion

• Genetic correlations between pure- and crossbreds
  for production traits are high
• Therefore, conventional testing and selecting of
  purebreds is validated by this study

18
Conclusion

• Genetic correlations between pure- and crossbreds
  for reproduction traits are low to moderate
• Therefore, we must include crossbred information
  into the genetic evaluation procedure to improve
  reproduction traits of crossbred sows

19
ThanksDept. Livestock Development
(DLD)ThailandUNERS / IPRSAGBU UNE
20
Genetic links b/w cross purebreds
51 common grandparents (21 grandsires 30
grandams)
481 Crossbred litters (38) (Total 1265)
544 Purebred litters (5) (Total 10558)
Litter records
21
Methods and Models

• Treat pure- and crossbred records as different
  traits eg. ADGp and ADGc
• Bivariate analysis of the two separate traits
  using ASReml animal model
• Estimate additive genetic covariances and genetic
  correlations between the two traits

22
Material and methods

• DU, LW, LR their crosses
• 11-year (1993-2003)
• 9075 performance test records
• 11823 litter records (from 3908 sows)

23
Number of breeds per contemporary group (HYS)
Test records
Litter records
Frequency ()
Frequency ()
Breeds/CG (fHYS)
Breeds/CG (tHYS)
24
Genetic links b/w cross purebreds
80 common parents (39 sires 41 dams)
429 Crossbred records (30) (Total 1409)
415 Purebred records (5) (Total 7666)
Performance test records