Inbreeding Using Genomics and How it Can Help Dr. Flavio S. Schenkel CGIL- University of Guelph
Introduction Why is inbreeding a concern? The biological risks of inbreeding: Inbreeding depression Accumulation of deleterious alleles Reduction of genetic variance Loss of genetic diversity
Inbreeding depression Trait Loss per 1% inbreeding Fat yield (kg) 1.1 Protein yield (kg) 0.5 Conformation (points) 0 Days open (days) 1.4 Calf survival 1 st calving (%) 0.5 Productive life (days) 13 (Van Doormaal, 2008. CDN report. March 2008)
Genetic variance Loss of genetic variance: e.g. Canadian Holstein 460,000 455,000 Additive genetic variance 450,000 Milk yield [kg 2 ] 445,000 440,000 435,000 430,000 425,000 420,000 415,000 410,000 1958 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 Year of birth (Stachowicz 2010. Ph.D. Thesis, UofG)
Genetic variance Loss of genetic variance: Genomic selection Change in genetic variance (%) after 10 generations Number of genes Selection scheme Few Many Genomic Selection -21.1-25.8 Traditional Selection -9.5-13.5 No selection (random) -5.4 0.2 (Adapted from Bastiaansen et al. 2012. Gen. Sel. Evol. 44:3)
Genetic diversity Loss (%) of genetic diversity: e.g. Canadian Holstein 0.07 0.06 6.5% Loss 0.05 0.04 0.03 0.02 0.01 0 1950 1958 1966 1974 1982 1990 1998 2006 Year (Stachowicz et al. 2011. JDS 94:5160 5175)
Introduction Inbreeding has been a concern for long time. Would genomic selection change this? No, it will be an even bigger concern. - Faster genetic progress associated with: Shorter generation interval Higher inbreeding rate per year No time for selection to counter balance negative effects of inbreeding
Faster genetic progress An illustration (Van Doormaal, 2012. CDN report. July 2012)
Consequences of genomic selection Selection Scheme Inbreeding per generation (%) Generation interval (years) Inbreeding per year (%) Genetic change per generation (%) Conventional 1.15 4.74 0.24 22.38 Genomic: Turbo 0.74 2.38 0.31 45.11 (Adapted from Buch et al. 2012. J. Anim. Breed. Gen. 129: 138 151)
Consequences of genomic selection At genomic level IBD_T F AVE IBD_T AVE IBD_Q QTL EFF P DIFF 0.25 1.00E+00 IBD / Inbreeding 0.2 0.15 0.1 0.05 8.00E-01 6.00E-01 4.00E-01 2.00E-01 QTL effect / Difference in allele frequency 0 1 1 2 3 3 4 5 6 7 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 25 26 28 29 Genome 0.00E+00 IBD_T= True IBD, F= Pedigree-based Inbreeding, AVE IBD_T= Average IBD_T, AVE IBD_Q= Average IBD at the QTLs, QTL EFF= QTL effect, P DIFF= Change in QTL allele frequency (Stachowicz, 2010. Ph.D. Thesis, UofG)
Average Inbreeding - All animals Introduction Is the observed rate of inbreeding going up with genomic selection? Yes, there could be some evidence already. E.g. Canadian Holstein Inbreeding coefficient 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 Year of birth Inbreeding coefficient Average Inbreeding - All animals 0.066 0.064 0.062 0.06 0.058 0.056 0.054 0.052 0.05 2008 2009 2010 2011 2012 Year of birth
Introduction Would an extra attention on inbreeding with genomics be needed? Yes - Breeding age of daughters of a bull and availability of semen from his sons are aligned. Risk of mating of siblings - Shorter generation interval, faster inbreeding rates. Co-selection of long haplotypes Higher accumulation of deleterious recessives
Co-selection of haplotypes Traditional Pedigree Sire of Sire Animal 1 Animal 2 Sire Dam Dam of Sire Sire of Dam Dam of Dam
Co-selection of haplotypes Genomic Pedigree ctgtagcgatcg atgtcgctcacg ctgtctagatcg atggatcgatcg ctgtagcgatcg cgatctagatcc agagatcgatcg atgtcgctcacg atagatcgatcg ctgtagcttagg agggcgcgcagt cgatctagatcc cggtagatcagt agagatcgatcg atggcgcgaacg ctatcgctcagg Higher selection pressure in parts of the genome (haplotypes)
Co-selection of haplotypes Measurement of inbreeding by chromosome segments instead of individual locus (From Pryce et al. 2012. JDS 95 :377 388)
Co-selection of haplotypes Individuals with similar genomic or pedigree relationships can have quite different proportions of chromosome segments that are identical by descent (IBD). If deleterious homozygotes are more likely to arise as a consequence of recent inbreeding, then strategies to minimize IBD chromosome segments could be a way of reducing them (Pryce et al. 2012).
Possibilities with genomics Assessment of inbreeding at genomic level - Under genomic selection, pedigree-based inbreeding does not reflect well the true genomic inbreeding. 0.25 IBD_T F AVE IBD_T AVE IBD_Q 0.2 IBD / Inbreeding 0.15 0.1 0.05 0 1 1 2 3 3 4 5 6 7 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 25 26 28 29 Genome
Possibilities with genomics - Under genomic selection controlling inbreeding using pedigree or genomic relationships has different effectiveness on genomic inbreeding. Target Genetic Pedigree Genomic Inbreeding Change Inbreeding Inbreeding Constraint on pedigree inbreeding 0.010 3.3 0.010 0.030 Constraint on genomic inbreeding 0.010 2.4 0.007 0.010 (Adapted from Sonesson et al. 2010. 9 th WCGALP)
Possibilities with genomics Controlling inbreeding using pedigree (P) or genomic (G) relationships Impact on the genome = Using P = Using G = Variance (From Sonesson et al. 2010. 9 th WCGALP)
Selection and Mating Systems Selection system: Truncated selection: TS (select animals with top evaluations as parents) Optimum selection: OS (selection based on both genetic progress and inbreeding) Mating system: Minimum Co-ancestry: MC (mate animals that are least related as possible) Random Mating: RM (mate animals at random)
Selection and Mating Systems Optimum selection (OS) vs. Truncated Selection (TS) Minimum Co-ancestry (MC) vs. Random mating (RM) Traditional evaluation 0.22 0.2 0.18 TS & RM TS & MC OS & RM OS & MC TR EBV RM TR EBV MC OC EBV RM OC EBV MC h 2 = 0.30 Inbreeding 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 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 25 26 27 28 29 30 Generation (Adapted from Stachowicz, 2010. Ph.D. thesis, UofG)
Selection and Mating Systems Optimum selection (OS) vs. Truncated Selection (TS) Minimum Co-ancestry (MC) vs. Random mating (RM) Genomic evaluation 0.22 0.2 0.18 TS & RM TS & MC OS & RM OS & MC TR GAS RM TR GAS MC OC GAS RM OC GAS MC h 2 = 0.30 Inbreeding 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 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 25 26 27 28 29 30 Generation (Adapted from Stachowicz, 2010. Ph.D. thesis, UofG)
Selection and Mating Systems Genome-wide true inbreeding after 30 generations 0.25 TR EBV Traditional RM Evaluation TR EBV MC TR GAS RM Genomic Evaluation TR GAS MC TS & RM TS & MC TS & RM TS & MC h 2 = 0.30 0.2 0.15 IBD 0.1 0.05 0 1 1 2 3 3 4 5 6 7 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 25 26 28 29 Genome (Adapted from Stachowicz, 2010. Ph.D. thesis, UofG)
Selection and Mating Systems Phenotypic gain after 30 generations TR TS h^2=0.30 OC OS h^2=0.30 6 5 Phenotypic values 4 3 2 1 0 EBV RM RM EBV MC MC GAS RM RM GAS MC MC Traditional Evaluation Genomic Evaluation (Adapted from Stachowicz, 2010. Ph.D. thesis, UofG)
Possibilities with genomics Discovery and test for unfavorable or deleterious recessive alleles/haplotypes: Examples: - E.g. CVM (Thomsen et al. 2006) - Fertility haplotypes (VanRaden et al. 2011)
Possibilities with genomics Haplotypes impacting fertility in Holstein Impact on Haplotype Freq CR NRR Earliest Known Ancestor(s) HH1 4.5% -3.1% -1.1% Pawnee Farm Arlinda Chief HH2 4.6% -3.0% -1.7% Willowholme Mark Anthony HH3 4.7% -3.2% -3.1% Gray View Skyliner Glendell Arlinda Chief (Adapted from VanRaden et al. 2011. JDS 94 :6153 6161)
Conclusions The availability of high-density SNP data has revolutionized dairy cattle breeding, changing breeding schemes and leading to higher rates of genetic gain. Higher rates of genetic gain from genomic selection will be associated with much higher annual rates of inbreeding, so methods to control inbreeding will become increasingly important.
Conclusions For traditional selection, pedigree-based inbreeding is an appropriate measure of true inbreeding across the genome. However, for genomic selection, inbreeding around selected markers can be much higher than the estimated pedigree-based inbreeding. This requires re-evaluate what is an appropriate rate of inbreeding when directly measured in the genome.
Conclusions Continued use of mating strategies, such minimum co-ancestry mating, is even more important under genomic selection. The application of optimum contribution selection could further help in controlling inbreeding. The use of genomic relationship information could better control inbreeding surrounding the markers.
Conclusions Genomic information can provide new tools for monitoring genome-wide inbreeding and controlling the frequency of deleterious recessive alleles/haplotypes.
Acknowledgments Ph.D. student: Katarzyna Stachowicz Pedigree data from: The Canadian Dairy Network (CDN) Financial support from: DairyGen Council of CDN Natural Sciences and Engineering Research Council of Canada (NSERC)
Thank you