200 7 Paul VanRaden, Curt Van Tassell, George Paul VanRaden, Curt Van Tassell, George Wiggans, Tad Sonstegard, and Jeff Wiggans, Tad Sonstegard, and Jeff O’Connell O’Connell Animal Improvement Programs Laboratory and Bovine Functional Genomics Laboratory, USDA Agricultural Research Service, Beltsville, MD, USA 200 8 Genomic Prediction Genomic Prediction Results Results
25
Embed
2007 Paul VanRaden, Curt Van Tassell, George Wiggans, Tad Sonstegard, and Jeff O’Connell Animal Improvement Programs Laboratory and Bovine Functional Genomics.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
2007
Paul VanRaden, Curt Van Tassell, George Wiggans, Paul VanRaden, Curt Van Tassell, George Wiggans, Tad Sonstegard, and Jeff O’ConnellTad Sonstegard, and Jeff O’Connell
Animal Improvement Programs Laboratory and Bovine Functional Genomics Laboratory, USDAAgricultural Research Service, Beltsville, MD, [email protected]
Genotyping and DNA extraction:• BFGL, U. Missouri, U. Alberta,
GeneSeek, GIFV, and Illumina
Computing from AIPL staff
NAAB DSEC April 2008 (10) Paul VanRaden200
8
Genomic MethodsGenomic Methods
Direct genomic evaluation• Inversion for linear prediction, REL• Iteration for nonlinear prediction
Combined genomic evaluation• Traditional PA or PTA, subset PA or
PTA, and direct genomic combined by REL in 3 x 3 selection index
• Nonlinear genomic predictions used
NAAB DSEC April 2008 (11) Paul VanRaden200
8
Nonlinear and Linear Regressions Nonlinear and Linear Regressions for marker allele effectsfor marker allele effects
NAAB DSEC April 2008 (12) Paul VanRaden200
8
Actual Results (Feb 2007 data)Actual Results (Feb 2007 data)
August 2003 PTAs for 2650 older bulls to predict January 2008 daughter deviations for 569 younger bulls (total = 3119 bulls)
Results computed for 27 traits: 5 yield, 5 health, 16 conformation, and Net Merit
Nonlinear A used, B didn’t work
NAAB DSEC April 2008 (13) Paul VanRaden200
8
Marker P-Values for Net MeritMarker P-Values for Net Merit
NAAB DSEC April 2008 (14) Paul VanRaden200
8
Marker Effects for Net MeritMarker Effects for Net Merit
NAAB DSEC April 2008 (15) Paul VanRaden200
8
Marker Effects for MilkMarker Effects for Milk
NAAB DSEC April 2008 (16) Paul VanRaden200
8
Marker Effects for Final ScoreMarker Effects for Final Score
NAAB DSEC April 2008 (17) Paul VanRaden200
8
Reliabilities and R-square values comparing Reliabilities and R-square values comparing traditional to genomic predictionstraditional to genomic predictions
Squared corr (x100)
Reliability
Traditional Genomic Genomic
Trait PA Genomic PA Realized Gain
Net Merit 8 21 36 54 18
Milk 30 44 38 54 16
Fat 15 39 38 65 27
Protein 31 43 38 51 13
Fat % 28 58 38 72 34
Protein % 32 51 38 66 28
NAAB DSEC April 2008 (18) Paul VanRaden200
8
Reliabilities and R-square values comparing Reliabilities and R-square values comparing traditional to genomic predictionstraditional to genomic predictions
Each animal has two evaluations• Expected genetic merit of daughters• Expected genetic merit of sons• Difference is sum of effects on X• SD = .1 σG, smaller than expected
Correlation with sire’s daughter vs. son PTA difference was significant (P<.0001), regression close to 1.0
NAAB DSEC April 2008 (23) Paul VanRaden200
8
SNP Density ComparisonSNP Density Comparison2130 older and 261 younger bulls2130 older and 261 younger bulls
REL of PA
Genomic REL
Trait 10K 20K 40K
Net Merit 35 45 49 48
Milk 37 49 52 51
Fat 37 52 55 57
Protein 37 52 54 53
Productive Life 28 44 45 42
SCS 31 44 45 47
Dtr Preg Rate 21 44 49 46
NAAB DSEC April 2008 (24) Paul VanRaden200
8
Genetic Evaluation AdvancesGenetic Evaluation Advancesand increases in genetic progressand increases in genetic progress
Year Advance % Gain
1935 Daughter-dam comparison 100
1962 Herdmate comparison 50
1974 Modified cont. comparison 5
1977 Protein evaluated 4
1973 Records in progress 10
1989 Animal model 4
1994 Net merit, PL, and SCS 50
2008 Genomic selection >40
NAAB DSEC April 2008 (25) Paul VanRaden200
8
ConclusionsConclusions
Genomic predictions significantly better than parent average (P < .0001) for all 26 traits tested
Gains in reliability from 2650 bulls (Feb data) equivalent on average to 9 daughters with records
April data included 5285 proven bulls, more analysis needed