Docility EPD: A Tool for Temperament Devori W. Beckman Iowa State University.

Docility EPD: A Tool for Temperament

Devori W. Beckman

Iowa State University

2007 NBCEC Brown Bagger SeriesOctober 24, 2007

What is Docility?

Temperament – reaction of beef cattle to handling by humans (Burrow, 1997)

Why Measure Docility? Ease of Handling ADG Carcass Quality RFI

Measuring Docility Flight Speed/Time, Exit Velocity Pen Score, Docility Test, Separation Test Chute/Crush Score

Docile Aggressive

Scoring System

1 2 3 4 5 6

BIF Guidelines (2002)

Docility EPD: Breed Associations North American Limousin Foundation (NALF)

1st Genetic Evaluation in 1998 Scores allocated at weaning Threshold (1, 2, 3-6); h2 = 0.40

American Angus Association (AAA) Genetic Evaluation Spring 2008

Scores allocated at yearling Threshold (1, 2, 3, 4-6); h2 = 0.37

Kuehn et al. (1998); Hyde (2003); Northcutt (2007)

Docility EPD: Sire Comparison Example

Bull Docility EPD

Odie + 20

Rowdy - 15

Difference = 35 Docility EPD measured on a percent basis.

Genetic Parameter Estimates For Docility In Limousin

Cattle

1. Maternal (M) Effects on Docility2. Sire by Herd (SH) Effects3. Further Investigation of M and SH Effects

Recent Work: Maternal Effects

Models logL h2D h2

M rDM C2

D -2569 .34 ± .01

DM -2559 .37 ± .03 .05 ± .02 -.41 ± .09

DM-Zero -2564 .31 ± .02 .02 ± .01

DC -2564 .31 ± .02 .03 ± .01

DMC -2554 .38 ± .03 .04 ± .03 -.55 ± .09 .04 ± .01

DMC-Zero -2562 .29 ± .02 .01 ± .01 .02 ± .01

Beckman et al. (2007)

Recent Work: Sire × Herd

Models logL h2D h2

M rDM C2 SH2

DMC -2554 .38 ± .03 .04 ± .03 -.55 ± .09 .04 ± .01

DMSH -2523 .16 ± .03 .02 ± .01 .35 ± .29 .06 ± .01

Notter et al. (1992); CSU - Unpublished Data

Objective

Do Maternal or SH effects appropriately describe docility, or are they artifacts of the nature of this data.

Previous Models Assumed Homogeneity

1 2 3 4 5 60%

5%

10%

15%

20%

25%

30%

35%

40%Docility Score Distribution

Tota

l Obs

erva

tions

, %

Beckman et al. (2007)

Additional Data Filtration

83%

17%

Distribution of Sires Across HerdsSingle Herd Multiple Herds

Additional Data Filtration

63%

37%

Distribution of ProgenySingle Herd Multiple Herds

Assessment of Heterogeneity

Fixed Effects

• WCG• AOD

Random Effects

• Direct• Maternal• SH• Residual

Step 1: Mixed Model (BASE) Analysis

Assessment of Heterogeneity Step 2: Fixed Model

Results: Fixed Model

Source df MS F-value P-value

Sire 262 0.251 2.68 < 0.0001

Herd 185 0.433 4.62 < 0.0001

SH 1160 0.102 1.09 0.027

Beckman and Garrick (2007)

Variance Component Estimation

Mixed Model (RESID) Analysis

|ê| = Xb + ZDuD + ZMuM + ZSHuSH + e

Results: Variance Components

Analysis h2D h2

M rDM SH2

BASE .18 ± .05 .03 ± .02 .13 ± .36 .04 ± .01

RESID .00 ± .00 .00 ± .00 .00 ± .00 .02 ± .01

Beckman and Garrick (2007)

Results: Variance Due to Herd

Conclusions

1. Maternal and SH Interaction Effects were Artifacts of the Data

2. Most of Heterogeneity Due to Herd Effects

Implications

Accuracy of Genetic Evaluation Total Herd Reporting

Garrick and Van Vleck (1987)

What’s Next?

Further Investigation of Heterogeneity

Clustering of Herds

Acknowledgements

North American Limousin Foundation American Angus Association Colorado State University Iowa State University

Thank you!

Image from schooldiscovery.com