Sire Fertility Genomics H Duane Norman Animal Improvement

Sire Fertility & Genomics H. Duane Norman Animal Improvement Programs Laboratory Agricultural Research Service, USDA, Beltsville, MD duane. norman@ars. usda. gov California Dairy Herd Improvement Association 1) ( 2008

Service-sire fertility –history l l Estimated relative conception rate(ERCR) w 70 -day nonreturn rate (NRR) w Source: − Dairy Records Management Systems, 1986− 2005 − USDA, 2006− 2008 Western Bull Fertility Analysis w 75 -d confirmed conception rate(CR) w Source: Agri. Tech Analytics, 2003−present California Dairy Herd Improvement Association 2) ( H. D. Norman 2008

Sire conception rate (SCR) l New USDA fertility evaluation w l l Initiated Aug. 2008 Based on confirmed Conception Rate Why is SCR most accurate? w Inseminations from~80% of DHI herds in US w Most services (not just first) w More effects accounted for California Dairy Herd Improvement Association 3) ( H. D. Norman 2008

Data included l Only AI inseminations with confirmed pregnancy l Service numbers 1– 7 for cows in lactations 1– 5 l Inseminations between 30 and 365 days in milk l Minimum standardized (ME) milk yield w >10, 000 lb for Holsteins w >8, 000 lb for Brown Swiss w >6, 000 lb for all other breeds California Dairy Herd Improvement Association 4) ( H. D. Norman 2008

Data included (cont. ) l l Most recent 4 yr. of inseminations The 6 traditional dairy breeds w Ayrshire w Brown Swiss w Guernsey w Holstein w Jersey w Milking Shorthorn California Dairy Herd Improvement Association 5) ( H. D. Norman 2008

Data excluded l Embryo-transfer donors l Services with sexed semen l Heifer services l If services within 10 d of each other, only the later of the 2 used California Dairy Herd Improvement Association 6) ( H. D. Norman 2008

Data excluded (cont. ) l Herd with ≥ 50% of milking cows without recorded breeding l Herd with average CR <10% or >90% l Service sire <0. 8 yr old l Services just prior to submission deadline (<70 d) California Dairy Herd Improvement Association 7) ( H. D. Norman 2008

Data sources l l 3 dairy records processing centers contributed >99% of data w Agri. Tech Analytics: California w Ag. Source Cooperative Service: Wisconsin w Dairy Records Management Systems: North Carolina 46 States and Puerto Rico California Dairy Herd Improvement Association 8) ( H. D. Norman 2008

Development of SCR l 4 -year research effort prior to implementation – primarily by Dr. Melvin Kuhn l Bull variables (expanded service-sire effects) l Cow variables (nuisance variables) California Dairy Herd Improvement Association 9) ( H. D. Norman 2008

Bull variables l Inbreeding coefficient of: w Service sire w Embryo l Bull age l Combined AI organization x mating year l Bull California Dairy Herd Improvement Association 10) ( H. D. Norman 2008

Cow variables l Combined herd, mating year, cow parity, and cow registry status l Combined mating month, year, and State l Cow parity l Service number l Short interval between matings l Cow age l Cow standardized milk yield l Cow’s permanent environment l Cow’s genetics California Dairy Herd Improvement Association 11) ( H. D. Norman 2008

SCR model Hoard’s Dairyman “The most complex model that I know of to evaluate animal performance” — Bennet Cassell, VPISU, 2008 California Dairy Herd Improvement Association 12) ( H. D. Norman 2008

Accuracy of SCR l l Reliability (Rel. ) = n/(n + 260) w n = number of inseminations w Constant 260 was needed based of variance components estimated for this model Confidence interval (CI) = w 0. 02313 = true standard deviation w 1. 282 = standard normal variate from normal distribution for an 80% CI California Dairy Herd Improvement Association 13) ( H. D. Norman 2008

Relationship of Rel. and 80% CI Inseminations 200 300 500 1, 000 2, 000 5, 000 10, 000 15, 000 20, 000 Rel. (%) 43 54 66 79 88 95 97 98 99 California Dairy Herd Improvement Association 14) ( 80% CI ± 2. 2 ± 2. 0 ± 1. 7 ± 1. 3 ± 1. 0 ± 0. 7 ± 0. 5 ± 0. 4 ± 0. 3 H. D. Norman 2008

SCR published l l Released 3 times a year with USDA genetic evaluation runs w January w April w August Eligible AI bulls w Active AI w Progeny test California Dairy Herd Improvement Association 15) ( H. D. Norman 2008

SCR published (cont. ) l l Overall matings w Holstein ≥ 300 in ≥ 10 herds w Ayrshire, Brown Swiss, ≥ 200 in ≥ 5 herds Guernsey, Jersey Matings during current 12 mo w Holsteins, Jersey ≥ 100 w Ayrshires, Brown Swiss, ≥ 30 Guernsey California Dairy Herd Improvement Association 16) ( H. D. Norman 2008

Interpretation of SCR l l l Predictor of bull fertility w Indicates Conception Rate w Reported as a percentage Average bulls SCR is 0. 0% Standard deviation for SCR is 2. 4%, of bulls between -2. 4% and +2. 4% California Dairy Herd Improvement Association 17) ( H. D. Norman 2/3 2008

Examples l l Bull with SCR of +3. 0% should provide a 3% higher CR than an average bull (SCR of 0. 0), and 6% higher CR than a bull with SCR of − 3. 0% Bull with SCR of +3. 0% expected to provide a CR of 33% in herd that has been averaging 30% and has been using “average” fertility bulls California Dairy Herd Improvement Association 18) ( H. D. Norman 2008

Herd & service sire fertility l l l Relationship between fertility of herd and bull SCR when examined together Herd-years stratified into 3 equally sized groups by CR w ≤ 27. 3% Low fertility w 27. 4 to 33. 9% Medium fertility w ≥ 34. 0% High fertility Bulls stratified into 3 equally sized groups by SCR ≤− 0. 9% w − 0. 8 to +1. 0% w ≥+1. 1% w California Dairy Herd Improvement Association 19) ( Low fertility Medium fertility High fertility H. D. Norman 2008

Herd CR (August 2008) Service-sire fertility Low Medium High Difference Low 20. 3 22. 6 24. 8 4. 5 California Dairy Herd Improvement Association 20) ( Herd fertility Medium 27. 4 30. 0 32. 4 High 35. 3 38. 7 41. 4 5. 0 6. 1 H. D. Norman 2008

Conclusions l SCR more accurate because it uses more inseminations w More DHI herds w Extra services (2– 7) w More complete model California Dairy Herd Improvement Association 21) ( H. D. Norman 2008

Genomics, genomics You 9 What? H. Duane Norman Animal Improvement Programs Laboratory Agricultural Research Service, USDA, Beltsville, MD duane. norman@ars. usda. gov California Dairy Herd Improvement Association 22) ( 2008

Genomic evaluations – history l l l Illumina Bovine. SNP 50 Bead. Chip developed (December 2007) Unofficial genomic evaluations for Holsteins initiated (April 2008); computed every 2 mo. w Owner letters and computer files distributed by AIPL Unofficial genomic evaluations for Jerseys (Oct. 2008) California Dairy Herd Improvement Association 23) ( H. D. Norman 2008

Genomic eval. – history (cont. ) l l Brown Swiss genomics tested; negotiations started to exchange Brown Swiss genotypes with Switzerland (Oct. 2008) Unofficial genomic evaluations for Brown Swiss (Dec. 2008) Over 22, 000 animals genotyped(Feb. 2009) Genomic evaluations become official(Jan. 2009) w Owner letters and computer files distributed by breed associations and NAAB California Dairy Herd Improvement Association 24) ( H. D. Norman 2008

Genomic evaluations: what’s it about? l l l DNA extracted from blood, hair, or semen ~40, 000 genetic markers (SNPs) tested simultaneously (about 1/2 cent per test) Value of each SNP determined by examining how each SNP impacts “tested animal” performance The payoff is to genotype young animals and apply the prediction equations Genomic evaluation combines SNP effects with traditional PA or PTA California Dairy Herd Improvement Association 25) ( H. D. Norman 2008

Genomic vs. traditional PTA l l Genotype is additional source of information (like parents, progeny, and animal records) For each animal, the genomic test is used to calculate genomic evaluations for all 29 traits Genomic evaluation interpreted the same as traditional PTA Expected to increase genetic improvement by 50% with decreased generation interval. Also, genomics contributes ~15 daughter equivalents to a bull California Dairy Herd Improvement Association 26) ( H. D. Norman 2008

Genomic evaluations – Jan. 2009 l l l Genomic evaluations became official Genotyped ancestors contribute their evaluations to descendents, not the reverse Evaluations of all genotyped females are public Evaluations of genotyped males either enrolled with NAAB or ≥ 24 mo. are public Young-bull genomic evaluations may be shared among AI organizations or disclosed by owner California Dairy Herd Improvement Association 27) ( H. D. Norman 2008

Genomic eval. – producer impact l l Young-bull genomic evaluations have accuracy approaching 1 st-crop daughter evaluations (60 -70% Reliability) AI organizations have started marketing genomically evaluated 2 -year-old bulls AI organizations are requiring genotypes on potential bull dams Progeny-test programs are changing California Dairy Herd Improvement Association 28) ( H. D. Norman 2008

Genomic evaluations – schedule l l Genomic evaluations provided at each of the 3 annual traditional genetic runs: January, April, and August Genomic evaluations on new bulls provided once or twice between traditional runs w We re-estimate SNP predictors when large numbers of predictor animals added California Dairy Herd Improvement Association 29) ( H. D. Norman 2008

Genomic eval. – increase accuracy? l l Genotyping more predictor bulls(most active-AI bulls expected to be genotyped soon) Should reach 1, 500 Brown Swiss through foreign collaboration Aggressive genotyped in domestic Jerseys underway, foreign collaboration is likely Investigate across-breed analysis to see if Holstein data helps accuracy for Jerseys and Brown Swiss California Dairy Herd Improvement Association 30) ( H. D. Norman 2008

How animals get genotyped l l Participating AI organizations have 5 -year exclusive right to evaluate bulls genomically Each AI organization genotypes 1 st-choice flushes, thus avoiding duplicate genotyping Web-based system being developed to show activity w Should help avoid expensive duplication Breed associations developing cow genotyping service California Dairy Herd Improvement Association 31) ( H. D. Norman 2008

On horizon: Low-cost genotyping l Developing a new genomic test, inexpensive enough to use for most animals l 384 SNPs proposed for new genomic test l Will provide parentage verification/discovery l Will provides a genetic estimate accurate enough for 1 st-stage screening California Dairy Herd Improvement Association 32) ( H. D. Norman 2008

Implications for dairy industry l l Rapid acceptance and use of genomic evaluations Young-bull acquisition and marketing now based on genomic evaluations Diversity of bull dams needs considered Industry groups taking responsibility for genotyping and validation California Dairy Herd Improvement Association 33) ( H. D. Norman 2008

International implications l l All major dairy countries reviewing genomic Interbull discussing how genomic evaluations should be integrated Balance needed between treating genotypes as proprietary versus open sharing Some importing countries might change rules to allow young genomic-tested bull use California Dairy Herd Improvement Association 34) ( H. D. Norman 2008

Fertility & genomics Overload ! H. Duane Norman Animal Improvement Programs Laboratory Agricultural Research Service, USDA, Beltsville, MD duane. norman@ars. usda. gov California Dairy Herd Improvement Association 35) ( 2008