Selection and Application – Making More Profitable Holsteins (male selection) North American Perspective Marjorie Faust and Katie Olson USA.

Selection and Application – Making More Profitable Holsteins (male selection)

North American Perspective

Marjorie Faust and Katie OlsonUSA

Making More Profitable Holsteins with Genomics What is the current state of the science?

How is the technology being applied and used in the US?

What are the challenges and opportunities for genomics going forward to ensure that Holsteins can deliver on producers’ demands for production efficiency & value for money?

Making and Delivering More Profitable Holsteins

Tremendous progress has been achieved in the US through industry cooperation and collaboration, pre-genomics and now into the genomic era– Milk recording organizations – DHI, Milk Testing Labs, DPRCs– Genetic Evaluation Unit – USDA-AIPL– Cooperative Extension– Breed Associations– Dairy Breeders– Universities– AI Organizations

-80

-60

-40

-20

0

20

40

1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Birth Year

Bree

ding

Val

ue -

Prot

ein,

kg

190

210

230

250

270

290

310

330

350

370

390

Prot

ein

Yiel

d, k

g

SiresCowsProtein Yield

Genomic Evaluations in North America

Released publicly in 2009 Rapid adoption into breeding programs

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

160,000

180,000

Jul-09 Jan-10 Aug-10 Feb-11 Sep-11 Apr-12 Oct-12 May-13

Genomic Evaluation, mo-yr

Num

ber G

enot

yped

, cum

ulati

ve

Total FemalesYoung FemalesTotal MalesYoung Males

USDA-AIPL, 2012

>68,000 males

>230,000 both sexes

Reliability for Genomic Evaluations in the US Young sires have always been a good source of

genetics and by 2012, genomics provide more accurate (but still imperfect) tools for identifying the best of the best

Trait Traditional GenomicGenomic-Traditional*

Protein Yield 35 75 +40Productive Life, mo. 26 72 +46Somatic Cell Score 30 76 +45Daughter Pregnancy Rate, % 26 71 +45Type Final Score 32 75 +42Calving Ease 33 57 +24

* Based on results from 44,950 Holstein young bulls.

Availability of gPTAs– Updated monthly– If male and <2 years of age, lists are not publicly available– Lists for proven bulls 3x per year

How are genomic evaluations being used?– Females

• 25-30% of genotypes are 50k or larger chip• Used for female culling and management• Used for breeding purposes, including bull-dam selection

– Males• ~90% of genotypes are 50k or larger chip• Use for screening is increasing

Application of Genomic Evaluations in the US

Young bulls as sires of sons– Pre-genomics ~10% or lower– By 2012, ~40% and growing– For bull calves born in 2011, sires are 23 months younger than for bull

calves born in 2006

Bull dams– Pre-genomics: Dominated by 2-year old cows and older– Post-genomics:

• Many more heifer contracts• More ownership of females by AI companies• More contracts to heifers out of young sires• More flushing of heifers, and IVF work with pre-puberty heifers• By 2012, genomic estimates drive value in elite breeding programs –

currently, the race is for the big numbers!

Impact of Genomics on Bull Selection Decisions

With the increase in the use of young bulls, there has been a concurrent increase in the number of sires of AI candidate bulls

Is this a true increase in sire families represented? Is the intent to find outcrosses or spread risk? Finding high ranking outcrosses still a challenge

Will semen sales also reflect an increase in sire families represented? Does this represent a real trend that will continue?

More Sires Represented in AI Candidate Bull Populations

126

82 82 89

0 2

62

187

020406080

100120140160180200

2008 2009 2010 2011Num

ber

of

Sire

s fo

r C

and

idat

e B

ulls

Birth Year of Young BullsUSDA-AIPL, 2012

Proven Bulls

Young Bulls

30-50% of inseminations to young bulls represent traditional Progeny Test Use of proven bulls has declined from ~70% to 50% of inseminations First crop bull usage has declined and been replaced by young bulls What will farmers demand? Where will young/proven bull usage

stabilize?

US Holstein Breeding Programs Using More Young SiresTo

tal I

nsem

inat

ion

s

Year of InseminationUSDA-AIPL, 2012

08

3642 48

2822

31

0

2

23

4039

38

54

29

4 310 11 15 13 1115

7 2 2 1

0%10%20%30%40%50%60%70%80%90%

100%

2007 2008 2009 2010 2011

Balancing the Economics of Increased Young Sire Usage

“Why would I pay $30 for older bulls when I can get high rankinggenomic young bulls for $15 per dose?” (US Dairy Breeder comment, Oct 2012)

– Semen price tends to be lower than proven bulls of equal genetic merit– Supply and demand should fix this but hasn’t done so in ~4 years

AI companies still working through strategies to produce young sire product– Young sires produce much lower volume of semen than proven bulls– Market life is shorter (no “branding” of bulls; faster turnover)– For same volume of semen sales, need more production stalls– More bulls required to produce product may reduce selection intensity

Acquisition cost per bull up ~50%-100% and climbing rapidly – More leases on extreme bulls (breeder retains ownership)– More bonus potential on extreme bulls

Production costs per dose of semen are considerably higher for young bulls

By 2010-2011, essentially all major US AI organizations scaledback on progeny test programs in one form or another– Several companies significantly reduced number of bulls sampled– Others reduced bull numbers and doses of progeny test semen

distributed per bull– Bull numbers appear to have rebounded recently, possibly to meet

market demands for more doses of product?

Collection of Breed Association linear type data on young sire progeny has evolved with genomics (Holstein Pulse, Summer 2011 and 2012)– >20% drop in the number of young sire daughters evaluated– Fewer sires represented, but many more progeny for some bulls

AI companies still progeny testing bulls, but this may change– Potential reduction in amount of phenotypic data (type)– Decision will largely be driven by farmer demand for proven bulls

Impact of Genomics on Progeny Test Programs

4 Years into Genomic Era – Is there value in progeny data?

460

480

500

520

540

560

580

600

April 2010 August 2012

Genomic Evaluation, mo-yr

Net

Mer

it, $

Young BullsProven Bulls

NM$ results for top 100 young and proven bulls from April 2010 Evaluation No change in NM$ formula over this time period For high ranking young bulls, PA & gPA overestimated daughter performance Progeny and performance data continue to add necessary information

High quality data are necessary for accurate, reliable evaluations. Current methodology requires unselected and unbia1sed data

As the industry continues to evolve, how do industry players share responsibility for ensuring that sufficient high quality data are collected?

The Challenge of Collecting Data in the Genomics Era ...“Fewer sires represented, but many more progeny for some bulls” (Holstein Pulse, Summer 2011 and 2012)

NameAug-09 TPISM

Aug-12 TPISM Herds Dtrs

Aug-09 PTAT

Aug-12 PTAT Dtrs-Type

PINE-TREE MARTHA SHOLTEN - ET 2216 1802 267 631 2.93 1.13 103MS ATLEES SHT AFTERSHOCK - ET 2270 1809 244 410 4.95 3.32 105GILLETTE JOB 2075 1746 240 349 3.14 1.4 293BADGER ONESHOT - ET 2069 1778 168 262 3.35 1.44 214LANGS-TWIN-B JANUARY - ET 2004 1830 124 248 2.92 1.72 135DE-SU GILLESPY - ET 2277 2157 144 227 4.15 2.68 176LADYS-MANOR RUBY D SHOUT - ET 2107 1920 111 201 3.42 2.06 159A-L-H ANTONIO - ET 1960 1665 83 168 3.55 1.75 77MOUNTFIELD MELVILLE - ET 1869 1802 58 96 2.46 1.72 68HILROSE FREELANCE TACOMA - ET 1611 1499 59 85 3.14 1.72 42

Challenges and Opportunities of Genomics for Inbreeding

Who is related to who? Success story for genomics! Who should be the parents of the next generation – balancing genetic

progress and inbreeding? Jury is still out on the impact of genomics. After selection, who should be mated to who? Genomic tools offer promise.

0

1

2

3

4

5

6

7

1950 1960 1970 1980 1990 2000 2010 2020

Birth Year

Inbr

eedi

ng, %

InbreedingExpected Future Inbreeding

1990

Across time, more traits and a wider range of traits included– Many newer traits are more challenging for making genetic progress

Broadening the Selection Goal for Holstein Profitability

Net Merit $1980

1980sTPI℠

1990sTPI℠

Today’sTPI℠

Production

Non-Production

Opportunity Presented by Low Heritability Traits

Traits with low heritability have always presented a challenge for making more profitable Holsteins

Making measurable progress quickly is more difficult because it is challenging to identify the best parents on a genetic level

-0.3

-0.25

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

1980 1985 1990 1995 2000 2005 2010 2015

Birth Year

Bree

ding

Val

ue -

SCS

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Som

atic

Cell

Scor

e

SiresCowsSCS

Challenge of Making Faster Genetic Progress

History teaches that even lowly heritable traits respond to intense selection on correlated traits – for example, genetic decline in fitness traits resulting from single trait selection on production traits

Are we measuring and monitoring all of the important/correlated traits?

-2.0

-1.0

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

1950 1960 1970 1980 1990 2000 2010 2020

Birth Year

Bree

ding

Val

ue -

DPR

, %

0

5

10

15

20

25

30

Dau

ghte

r Pre

gnan

cy R

ate,

%

CowsSiresDaughter Pregnancy Rate

Challenge and Opportunity of Genomics for New Traits

What traits need to be protected when selecting for traits like efficiency? Think of genomics as providing a Formula 1 race car. Driving a Formula 1 race

car can get us to the destination faster, but a crash at 150 miles/hour is considerably more painful than a crash at 20 miles/hour

Efficiency

Intake

Production, fertility, growth, immunity, etc.

As an industry, how do we share the responsibility of investingin the collection of high quality, unselected and unbiased data?

– Necessary to recalibrate SNP effects (SNP key)– Minimize time gap between real data and selection candidates– Include data on families that are most relevant to the candidate group– Needed for selection traits and correlated traits which need to be

protected

Methodology solutions needed in next steps of genomics developments– Improved comparability across age groups for both sexes

• Rankings within age groups do not appear to be biased, but more difficult to compare genetic rankings across age groups

– Dampening or minimizing the impact of potential bias from female side– Avoiding bias which arises from genomic pre-selection– Need the help of genetic evaluation units and academia

Challenges and Opportunities in the Genomic Era

Genomics is providing more accurate (but still imperfect) toolsfor identifying the best of the best at a younger age

As genomic tools have been improved, more emphasis on younger and younger parents and grandparents of both sexes– Seeing use of young sires pre-release in IVF done with pre-puberty

heifers out of young sires (3 generations removed from performance data)

– For breeding programs, less concern with reliability than in the past

Genetic diversity continues to require attention for both sexes– Genomics can be both beneficial and detrimental for genetic diversity– Must protect against real risks of increasing inbreeding via genomics

Continued methodology improvements and developments needed to ensure that genomic technologies remain both accurate and useful

Conclusions

For AI companies, lots of cost going in, but not reflected inincreased revenue (yet)– Semen price tends to be lower than proven bulls of equal genetic merit– Current option for AI companies is to cut cost. No PT. Less data

Driving the Formula 1 race car of genomics can get us to the destination faster, but a crash at 150 miles/hour is considerably more painful than a crash at 20 miles/hour– Progeny and performance data continue to add necessary information– The need to invest in phenotypic data is not shrinking, but growing and

expanding

Producers demand production efficiency & value for money, and the industry shared responsibility is to make sure that Holsteins can deliver!

Conclusions