current status of genomic evaluation for u.s. dairy cattle

Wiggans, 2013China Emerging Markets Program Seminar

Dr. George R. WiggansAnimal Improvement Programs LaboratoryAgricultural Research Service, USDABeltsville, MD 20705-2350301-504-8407 (voice) 301-504-8092 (fax)[email protected]

Current statusof genomic evaluation forU.S. dairy cattle

mailto:[email protected]


Genotypes received (last 12 months)

Breed Female MaleAll

animalsFemale:

maleAyrshire 515 337 852 60:40Brown Swiss 701 606 1,307 54:46Holstein 149,712 28,191 177,903 84:16Jersey 18,600 3,059 21,659 86:14All 169,528 32,193 201,721 84:16


Genomic data flow

DNA samples

genotypes

genomic

evaluations

nominati

ons,

pedigr

ee dat

a

genotype

quality reportsge

nomic

evalu

ations

DNA sam

ples

genotypes

DNA samples

Dairy Herd Improvement (DHI) producer

Council on Dairy Cattle Breeding (CDCB)

DNA laboratory AI organization,breed association


Evaluation flow

Animal nominated for genomic evaluation by breed association or AI organization

Hair or other DNA source sent to genotyping lab

Source Samples (no.) Samples (%)Blood 27,043 14Hair 116,833 59Nasal swab 4,619 2Semen 4,119 2Tissue 45,018 23


Evaluation flow (continued)

DNA extracted and placed on chip for 3-day genotyping process

Genotypes sent fromgenotyping lab to CDCB for accuracy review


Laboratory quality control

Each SNP evaluated for Call rate Portion heterozygous Parent-progeny conflicts

Clustering investigated if SNP exceeds limits

Number of failing SNPs indicates genotype quality

Target of <10 SNPs in each category


Before clustering adjustment

86% call rate


After clustering adjustment

100% call rate



Genotype calls modified as necessary

Genotypes loaded into database

Nominators receive reports of parentage and other conflicts

Pedigree or animal assignments corrected

Genotypes extracted and imputed to 45K


Imputation

Based on splitting genotype into individual chromosomes (maternal and paternal contributions)

Missing SNPs assigned by tracking inheritance from ancestors and descendants

Imputed dams increase predictor population

Genotypes from all chips merged by imputing SNPs not present


findhap

Developed by Dr. Paul VanRaden, ARS, USDA

Divides chromosomes into segments

Allows for successively shorter segments (usually 3 runs) Long segments lock in identical by descent Shorter segments fill in missing SNPs

Separates genotype into maternal and paternal contribution, haplotypes (phasing)

Builds haplotype library sequenced by frequency



SNP effects estimated

Final evaluations calculated

Evaluations released to dairy industry Download from CDCB FTP site with

separate files for each nominator Monthly release for new animals All genomic evaluations updated

3 times each year with traditional evaluations


Information sources for evaluations

Traditional evaluations of genotyped bulls and cows used to estimate SNP effects

Combined final evaluation Sum of SNP effects for an animal’s alleles Polygenetic effect Traditional evaluation

Pedigree data used and validated by genotypes


Genotypes evaluated

Jun A O Jan F A M J J A S O N D Jan F M A M J J A S O N D Jan F M A M J J A S O N D Jan F M A M J J A S0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000Young imputed

Old imputed

Female Young <50K

Male Young <50K

Female Old <50K

Male Old <50K

Female Young >=50K

Male Young >=50K

Female Old >=50K

Male Old >=50K

Evaluation date

Ani

mal

s ge

noty

ped

(no.

)

Imputed, youngImputed, old (young cows included before March 2012) <50K, young, female<50K, young, male<50K, old, female<50K, old, male ( 20 bulls)50K, young, female50K, young, male50K, old, female50K, old, male

2013201220112009 2010


Holstein prediction accuracy

*2013 deregressed value – 2009 genomic evaluation

Trait Bias* Reliability (%)Reliability gain

(% points)Milk (kg)

−80.369.2 30.3

Fat (kg)−1.4

68.4 29.5

Protein (kg)−0.9

60.9 22.6

Fat (%) 0.0 93.7 54.8Protein (%) 0.0 86.3 48.0Productive life (mo)

−0.773.7 41.6

Somatic cell score 0.0 64.9 29.3Daughter pregnancy rate (%) 0.2 53.5 20.9Sire calving ease 0.6 45.8 19.6Daughter calving ease

−1.844.2 22.4

Sire stillbirth rate 0.2 28.2 5.9Daughter stillbirth rate 0.1 37.6 17.9


Holstein prediction accuracy

*2013 deregressed value – 2009 genomic evaluation

Trait Bias* Reliability (%)Reliability gain

(% points)Final score 0.1 58.8 22.7Stature

−0.268.5 30.6

Dairy form−0.2

71.8 34.5

Rump angle 0.0 70.2 34.7Rump width

−0.265.0 28.1

Feed and legs 0.2 44.0 12.8Fore udder attachment

−0.270.4 33.1

Rear udder height −0.1

59.4 22.2

Udder depth −0.3

75.3 37.7

Udder cleft−0.2

62.1 25.1

Front teat placement −0.2

69.9 32.6

Teat length−0.1

66.7 29.4


Genotypes by animal age (last 12 months)

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24-35

36-47

48-59

60

0

5,000

10,000

15,000

20,000

25,000

30,000 Holstein male Holstein female Jersey male Jersey female

Age (mo)

Freq

uenc

y (n

o)


Parent ages for marketed Holstein bulls

Birth year

Dam

0

10

20

30

40

50

60

70

80

90

100

2007 2008 2009 2010 2011 2012

Pare

nt a

ge (m

o)

Sire


Marketed Holstein bulls

Year entered

AI

Traditional progeny-

testedYoung

genotypedAll

bulls2008 1,798 0 1,7982009 1,909 337 2,2462010 1,827 376 2,2032011 1,441 467 1,9082012 1,376 555 1,931


Genetic merit of marketed Holstein bulls

00 01 02 03 04 05 06 07 08 09 10 11 12-100

0

100

200

300

400

500

600

Year entered AI

Aver

age

net m

erit

($)

Average gain:$20.21/year




0 1 2 3 4 5

Genomic prediction of progeny test

Select parents, transfer embryos

to recipients

Calves born and

DNA tested

Calves born from DNA-selected

parents

Bull receives progeny test

Reduce generation interval from 5 to 2 years


Benefit of genomics

Determine value of bull at birth

Increase accuracy of selection

Reduce generation interval

Increase selection intensity

Increase rate of genetic gain


Why genomics works for dairy cattle

Extensive historical data available

Well-developed genetic evaluation program

Widespread use of AI sires

Progeny-test programs

High-value animals worth the cost of genotyping

Long generation interval that can be reduced substantially by genomics


Current organizational roles

Council on Dairy Cattle Breeding (CDCB) responsible for receiving data, computing, and delivering U.S. genetic evaluations for dairy cattle

Animal Improvement Programs Laboratory (AIPL) responsible for research and development to improve the evaluation system

CDCB and USDA employees co-located in Beltsville


Funding

CDCB evaluation calculation and dissemination funded by fee system Based on animals genotyped About 80% of revenue from bulls Higher fees for herds that

contribute less information

AIPL research on evaluation methodology funded by U.S. Federal government

$


Ways to increase accuracy

Automatic addition of traditional evaluations of genotyped bulls when bull is 5 years old

Possible genotyping of 10,000 bulls with semen in repository

Collaboration with other countries

Use of more SNPs from HD chips

Full sequencing (identify causative mutations)


Evaluation accuracy by included SNPs

*Difference in reliability from 45K in parentheses

Reliability (%)*Trait 45KMilk 69.2Fat 68.4

Protein 60.9Fat percentage 93.7Protein percentage 86.3Net merit 51.6Productive life 73.7Somatic cell score 64.9Daughter pregnancy rate 53.4Service-sire calving ease 45.8Daughter calving ease 44.2Service-sire stillbirth rate 28.2Daughter stillbirth rate 37.6

60K69.3 (0.1)68.7 (0.3)60.8 (–0.1)94.4 (0.7)87.1 (0.8)51.7 (0.1)74.0 (0.3)65.8 (0.9)54.1 (0.7)45.7 (–0.1)45.8 (1.6)28.3 (0.1)37.8 (0.2)

75K68.9 (–0.3)68.6 (0.2)60.6 (–0.3)93.9 (0.2)86.3 (0.0)51.6 (0.0)73.1 (–0.6)65.6 (0.7)53.6 (0.2)45.1 (–0.7)44.9 (0.7)28.7 (0.5)37.1 (–0.5)

91K69.2 (0.0)68.4 (0.0)60.8 (–0.1)93.5 (–0.2)86.1 (–0.2)51.3 (–0.3)73.8 (0.1)65.6 (0.7)53.8 (0.4)46.2 (0.4)44.9 (0.7)29.9 (1.7)39.2 (1.6)


Evaluation accuracy (continued)

*Difference in reliability from 45K in parentheses

Reliability (%)*Trait 45KFinal score 58.8Stature 68.5Dairy form 71.8Rump angle 70.2Rump width 65.0Feet and legs 44.0Fore udder attachment 70.4Rear udder height 59.4Udder depth 75.3Udder cleft 62.1Front teat placement 69.9Teat length 66.7

60K58.7 (–0.1)69.0 (0.5)72.2 (0.4)70.9 (0.7)65.4 (0.4)45.1 (1.1)70.6 (0.2)59.9 (0.5)76.2 (0.9)62.2 (0.1)70.1 (0.2)67.2 (0.5)

75K58.4 (–0.4)68.8 (0.3)71.9 (0.1)70.7 (0.5)65.0 (0.0)45.1 (1.1)70.0 (–0.4)59.6 (0.2)76.0 (0.7)62.0 (–0.1)70.2 (0.3)66.6 (–0.1)

91K58.7 (–0.1)69.1 (0.6)72.0 (0.2)70.9 (0.7)65.2 (0.2)45.1 (1.1)70.4 (0.0)59.8 (0.4)76.1 (0.8)62.2 (0.1)70.4 (0.5)66.9 (0.2)

All production, type, and fitness traits (0.5) (0.1) (0.4)


Key issues for the dairy industry

Inbreeding and genetic diversity(including across breeds)

Sequencing, new genes, and mutations

Novel traits, resource populations(feed efficiency, health, milk properties)


Application to more traits

Animal’s genotype good for all traits

Traditional evaluations required for accurate estimates of SNP effects

Traditional evaluations not currently available for heat tolerance or feed efficiency

Research populations could provide data for traits that are expensive to measure

Will resulting evaluations work in target population?


What’s already planned

Genomic evaluations for new traits Health (e.g., resistance to heat stress) Feed efficiency

Genomic mating programs Selection of favorable minor alleles Reduction of genomic inbreeding

Genomic evaluations based on more SNPs (60K)

Adding SNPs for causative genetic variants


What’s already planned (continued)

BARD project (Volcani Center, Israel) A priori granddaughter design (APGD) Identification of causative variants for

economically important traits

International collaboration on sequencing United States, United Kingdom, Italy, Canada Bulls selected using APGD


Parentage validation and discovery

Parent-progeny conflicts detected Animal checked against all other genotypes Reported to breeds and requesters Correct sire usually detected

Maternal grandsire (MGS) checking SNP at a time checking Haplotype checking more accurate

Breeds moving to accept SNPs in place of microsatellites


MGS detection — HAP method

Based on common haplotypes

After imputation of all loci, determine maternal contribution by removing paternal haplotype

Count maternal haplotypes in common with MGS

Remove haplotypes from MGS and check remaining against maternal great-grandsire (MGGS)


MGS detection — SNP method

Based on SNP conflicts

Check if animal and MGS have opposite homozygotes (duo test)

If sire is genotyped, some heterozygous SNPs can be checked (trio test)


MGS detection by breed

Ancestors confirmed (%)SNP method HAP method

Breed MGS MGS MGGSBrown Swiss 94 97 85Holstein 95 97 92Jersey 91 95 95


Haplotypes affecting fertility

Rapid discovery of new recessive defects Large numbers of genotyped animals Affordable DNA sequencing

Determination of haplotype location Significant number of homozygous animals

expected, but none observed Narrow suspect region with fine mapping Use sequence data to find causative mutation


Haplotypes affecting fertility

*Causative mutation known

Name

BTAchromo-

someLocation*

(Mbp)

Carrierfrequency

(%) Earliest known ancestorHH1 5 63.2* 4.5 Pawnee Farm Arlinda ChiefHH2 1 94.9–96.6 4.6 Willowholme Mark AnthonyHH3 8 95.4* 4.7 Glendell Arlinda Chief,

Gray View SkylinerHH4 1 1.3* 0.7 Besne BuckHH5 9 92.4–93.9 4.4 Thornlea Texal SupremeJH1 15 15.7* 23.4 Observer Chocolate SoldierBH1 7 42.8–47.0 14.0 West Lawn Stretch ImproverBH2 19 10.6–11.7 15.4 Rancho Rustic My DesignAH1 17 65.9–66.2 23.6 Selwood Betty’s Commander


Haplotype tracking of known recessives

*Causative mutation known

Recessive Haplotype

BTAchromo-

some

Testedanimals

(no.)Concord-ance (%)

New carriers

(no.)BLAD HHB 1* 11,782 99.9 314CVM HHC 3* 13,226 — 2,716DUMPS HHD 1* 3,242 100.0 3Mule foot HHM 15* 87 97.7 120Polled HHP 1 345 — 2,050Red coat color HHR 18* 4,137 — 5,927SDM BHD 11* 108 94.4 108SMA BHM 24* 568 98.1 111Weaver BHW 4 163 96.3 32


Progression of chips

2008 2009 2010

Official 3Kevaluations

DecUnofficial 3K

evaluations

Sep

Bovine3K BeadChip

(3K)Jul

BovineHD BeadChip

(777K)Jan

Official 50K Brown Swiss evaluations

AugOfficial 50K

Holstein & Jersey evaluations

JanUnofficial 50K

evaluations

Apr

BovineSNP50 BeadChip

(50K)Jan

2011 2012 2013

Official 12K evaluations

Oct

Zoetis LD BeadChip (12K)Sep

GGP v2 BeadChip

(19K)May


MayOfficial 77K evaluations

Jan

GGP HD BeadChip

(77K)Dec


Mar

GeneSeek Genomic Profiler (GGP) BeadChip (8K)Feb

Official7K & 648K

evaluations

Dec

BovineLD BeadChip

(7K)Sep


Aug

Affymetrix BOS 1 Plate Array

(648K)Jan


International dairy breeding

Genotype alliances North America (US, Canada, UK, Italy) Ireland, New Zealand Netherlands, Australia Eurogenomics (Denmark/Sweden/Finland, France,

Germany, Netherlands/Belgium, Spain, Poland)

Interbull genomic multitrait across-country evaluation (GMACE)


GMACE reference populations (August)

Country Animals (no.)Australia 5,314Denmark/Finland/Sweden 23,961France 24,313Germany 25,624Italy 21,041Netherlands 23,047Poland 3,174Switzerland (Red Holstein) 4,194


Impact on breeders

Haplotype and gene tests in selection and mating programs

Trend towards a small number of elite breeders that are investing heavily in genomics

About 30% of young males genotypeddirectly by breeders since April 2013

Prices for top genomic heifers can bevery high (e.g., $265,000 )


Impact on dairy producers

General

Reduced generation interval

Increased rate of genetic gain

More inbreeding/homozygosity?


Impact on dairy producers (continued)

Sires

Higher average genetic merit of available bulls

More rapid increase in genetic merit for all traits

Larger choice of bulls in terms of traits and semen price

Greater use of young bulls


Conclusions

Genomic evaluation has dramatically changed dairy cattle breeding

Rate of gain is increasing primarily because of a large reduction in generation interval

Genomic research is ongoing Detect causative genetic variants Find more haplotypes affecting fertility Improve accuracy through more SNPs, more

predictor animals, and more traits


U.S. genomic evaluation team

current status of genomic evaluation for u.s. dairy cattle

Documents