agriculture and agri-food canada agriculture et agroalimentaire canada the genetics of feed...

Agriculture andAgri-Food Canada

Agriculture etAgroalimentaire Canada

The Genetics of Feed Efficiency in CattleThe Genetics of Feed Efficiency in Cattle

Dr. D.H. “Denny” Crews, Jr.Research Scientist, Beef Quantitative GenomicsNational Study Leader, Livestock Genetics & GenomicsAAFC Research Centre, Lethbridge, Alberta



Many Measures of EfficiencyMany Measures of Efficiency

• Probably two dozen measures of efficiency have been described in beef cattle

• Feed conversion ratio is a gross measure of efficiency– Genetic trend has been positive along with growth

• Rg (FCR, growth): -0.61 to -0.95

– Related to increased mature weights and therefore, maintenance energy requirements

– Lends poorly to selection• Most selection pressure on growth rate



Reducing InputsReducing Inputs

• Very little genetic improvement has been aimed at reducing input costs:– Feed costs are the largest non-fixed cost of beef production– >70% of total variable costs

• Daily feed intake (dDMI) is heritable (h2 = 0.34 based on 23 studies [Koots et al., 1994a]) and therefore likely to respond to selection

Trait Rg (dDMI) Reference

WT205 0.67

Koots et al. (1994b)

WT365 0.79

MKTWT 0.92

BWT 0.77



Reducing Inputs: Feed EfficiencyReducing Inputs: Feed Efficiency

• Gross efficiency (Archer et al., 1999; gain/feed) and feed conversion ratio (FCR, feed/gain) have been discussed for more than 30 years, along with at least 20 other so-called efficiency measurements

• Most have at least moderate heritability (> 0.32 - 0.37) and strong genetic correlation with growth

ΔGFCR | WWT = (RgFCR,WWT) (h2 WWT) (i WWT) (σg(FCR)) = -0.21 kgd-1/gen



Selection: FCRSelection: FCR

• Adding feed conversion ratio to breeding objectives would have the following implications:– Additional ΔG for growth; the most immediate concern is

that with mature size (RgFCR,MWT > 0.50)

– Disproportionate selection on dDMI versus ADG. Gunsett (1984) discussed the problems associated with selection for ratio traits

– Negative genetic trend in FCR does not translate to incremental improvement in feed efficiency

• Changes in FCR can be made without changing efficiency (+ ADG)

• Selection response is usually unpredictable (Gunsett, 1984)



RFI DefinitionRFI Definition

• Residual feed intake (syn. net feed efficiency) is defined as the difference between actual feed intake and that predicted by regression accounting for requirements of production and body weight maintenance– dDMI = CG + ADG + BWT + “other production” + RFI– Regression can be either phenotypic or genetic– “Forced” independence with growth rate, stage of

production and weight alleviates problems with correlated response

– RFI phenotypes are independent of age, stage of production, and previous plane of nutrition



0.8 1 1.2 1.4 1.6 1.8 2 2.2

Average daily gain, kg/day

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

Residual feed intake, kg/day

r squared=0.000, n=75, P<0.9969

Figure 2. Relationship between residual feed intake and average daily gain.

60 70 80 90 100 110 120

Metabolic mid-weight, kg

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

Residual feed intake, kg/day

r squared=0.000, n=75, P=0.9936

Figure 1. Relationship between residual feed intake and metabolic body weight

Figures courtesy J. A. Basarab



An Expensive PhenotypeAn Expensive Phenotype

• Cost of data collection is high– $150-175 per head for equipment alone

• Intensive 70-90 d test period– Limited numbers of animals with phenotypes

• Technology is still developing– Reduction in altered feeding behavior: Individual

intake on group-fed cattle

• Commercial test facilities largely unavailable



Potential ReturnsPotential Returns• Most agree RFI is moderately heritable (0.30 to 0.40)• Can force independence with any production trait

– Typical RFI generally uncorrelated with body composition

• Preliminary research reports– Uncorrelated with mature size– Highly positive genetic correlation with mature cow

efficiency– No evidence of antagonism with reproductive merit

• Phenotypic and genetic variance– 5-7 lb per day phenotypic difference among yearling bulls– Similar variability among crossbred steers during finishing



Differences in RFI groupsDifferences in RFI groups

Crews et al., 2003

Intake per d, lb 21.7 25.5 <0.001

Total 84-d intake, lb 2121.9 2511.0 <0.001

Feeding Events per d 5.56 6.22 <0.001

Carcass fat, in 0.28 0.30 <0.110

Lean Yield, % 59.93 59.47 >0.240

Marbling score Select 80 Select 75 >0.640

RFI < 0.00 RFI > 0.00

Efficient Inefficient P-value

Total 84-d gain, lb 264 270 >0.470



Potential Industry ImpactPotential Industry Impact

• Our results show that the more efficient half of steers gained the same amount of weight, produced carcasses with the same yield and quality grades with the same amount of time on feed but consumed 390 pounds less feed than the less efficient half.

• In a region with 2+ million head processed per year, that 780 million pounds of feed costs almost $40 million.



RFI Genetic VariabilityRFI Genetic Variability

• Several studies have estimated genetic variance and heritability for RFI

Var(G) h2 Reference

0.28 Koch et al. (1963)

0.14 Fan et al. (1995)*

0.44 Arthur et al. (1997)

0.149 0.39 Arthur et al. (2001a)

0.220 0.39 Arthur et al. (2001b)

0.267 0.30 Crews et al. (2003a,b)



RFI Adjusted for Body CompositionRFI Adjusted for Body Composition

Trait % of DFI variance Rank Correlation, RFI-1

MWT + ADG 67.9 + 8.6 1.00

Gain in Empty Body Fat 3.9 0.92

Gain in Empty Body Water 1.1 0.90

Basarab et al. (2003)

• Adding gain in RTU rib fat and(or) RTU intramuscular fat provided similarly small increases in model R2



RFI Genetic CorrelationsRFI Genetic Correlations

Trait Rg(RFIp) Reference

Feed Conversion Ratio 0.70 Herd and Bishop (2000)Arthur et al. (2001a,b)Feed Conversion Ratio 0.85

Feed Intake 0.64

Feed Intake 0.79

Back Fat 0.17

Live weight 0.32 Arthur et al. (2001b)

ADG 0.10

Carcass REA -0.17 Schenkel et al. (2004)Crews et al. (2003a)

Carcass marbling score -0.44



Phenotypic Regression RFI and ProductionPhenotypic Regression RFI and Production

• RFI is defined as the component of feed intake that is phenotypically independent of production

• Recent studies have shown significant non-zero genetic correlation of RFIp with production, body weight, etc.

• RFIp usually contains a genetic component due to production

• The phenotypic variance of RFIp is completely described by– Heritability of feed intake and production– Genetic and environmental correlations of feed intake with

production– (Kennedy et al., 1993)



Repeatability of RFIpRepeatability of RFIp

• Archer et al. (2002) measured intake and derived RFIp on heifers postweaning and then on open cows following weaning of their second calf

• dDMI, ADG, MWT, FCR and RFI considered different traits between cows and heifers to estimate genetic correlations

• Rg > 0.85 strongly indicates genetic equivalence:

Trait Rg (cow x heifer)

DFI 0.94

ADG 0.72

MWT 0.82

FCR 0.20

RFIp 0.98



RFI and Multiple Trait SelectionRFI and Multiple Trait Selection

• Single trait selection is not advisable• Few attempts have been made to

incorporate RFI into selection schemes• An example multiple trait index was

developed by Crews et al. (2006)



Index ValuesIndex Values

0.342

0.066

1791.2-1.127

0.000

6.048

b =

SYM -1 0.156

0.017

709.48.664

0.049

2.019

0.000

3.708

0.000

-0.27

183.73

-21.49

-0.09

24.79

-10.12

=

I = -10.12 (RFI) + 24.79 (ADG) – 0.09 (YWT) ~ N ( 100 , 7.812 ; range: 80.1 – 115.7)



Correlations of Index Value with Component TraitsCorrelations of Index Value with Component Traits

RFI

dDMI

ADG

Bull Trait P - value (I)

YWT

YSC

0.01

0.03

0.01

0.90

0.12

-0.74

-0.22

0.53

0.01

0.16



SummarySummary

• RFI may be a candidate for genetic evaluation and improvement systems

• Independence with growth, body weight, and any identifiable source of dDMI covariance can be forced

• Heritability is at least as high as early growth but genetic variance is limited– Probably enough to make substantial economic

improvement

• Multiple trait selection schemes still required



SummarySummary

“Genetic improvement in efficiency of feed utilization is higher-hanging fruit”

John Pollak, BIF 2002



Thank you

[email protected]

agriculture and agri-food canada agriculture et agroalimentaire canada the genetics of feed...

Documents

genetics of feed efficiency

net feed efficiency

feed efficiency changes

feed costs

feed conversion ratio

growth rg fcr

growth rate slide

actual feed intake