ans3384selection - Selection: Its Effects on Animal...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Selection: Its Effects on Animal Performance ANS 3384 “Well, scratch No. 24. He did pretty good though – right up to the jet engine test!” Selection • Natural Selection – Survival of the Fittest – Result: Changes in gene frequencies in favor of the those genes resulting in higher reproductive/survival rates • Artificial Selection – Culling of those animals least desirable for economic traits – Result: Changes in gene frequencies in favor of those genes which improve economically important traits Artificial Selection • How do we pick animals that are going to improve the herd? • Animals that are going to improve the herd are genetically superior for ADDITIVE genes • Additive genetic superiority is passed on from parent to progeny • Those animals with higher BREEDING VALUES will produce superior progeny! BREEDING VALUES • Definition of a Breeding Value: – The Difference of the Additive Genetic Value of a Particular Animal from the Mean Additive Genetic Value (AGV) of the population/breed/herd • BV = Animal’s AGV – Average-AGV • We can never know an animal’s BV exactly but must predict it based on its performance/phenotype Predicting Breeding Values “Estimated Breeding Values” • H = Predicted AGV of an individual • H = Mean G + by.x (P – Mean P) – where P is performance/phenotype • H – Mean G = bg.p (P – Mean P) • EBV = bg.p (P – Mean P) • Now bg.p can equal h2 so ……. • EBV = h2 ( Performance – Mean Performance) Example of Calculating an Estimated Breeding Value • The yearling weight of a particular Angus bull, No. 234 is 1200 pounds • The mean yearling weight of the other bulls fed out in the same group at the same time (referred to as his CONTEMPORARIES) is 1000 pounds • If h2 = 0.40 • Thus, EBV234 = 0.40 (1200 – 1000) = 80 Expected Genetic Progress = ∆G • ∆G = h2 (Mean of Selected Parents – Mean of Parent’s Contemporaries) • SELECTION DIFFERENTIAL = – (Mean of Selected Parents – Mean of Parent’s Contemporaries) • EXAMPLE – SELECTION FOR WEANING WEIGHT – SELECTED BULL AND HEIFER CALVES’ WWT = 600 # – CONTEMPORARIES = 500 # – h2 = 0.30 • ∆G = 0.30 (600 – 500) = .30 (100) = 30 LBS. Selection Differential • The Selection Differential is directly related to proportion of the population kept for breeding purposes • In all livestock species fewer males are used for breeding than are females • One bull, even by natural service can breed 25 or more cows • Thus we need no more than 5% of the beef breed bull calves and perhaps as many as 50% of the heifers Selection Differential • As a result of the selection of fewer males, and the selection of the best animals of the males available, the selection differential of males is much higher than that of the females • Using the weaning weight example, the best 5% of the bull calves may weigh 700 lbs (a selection differential of 200 lbs) • The best 50% of the heifer calves, on the other hand, may weigh 550 lbs (a selection differential of 50 lbs) More on Expected Genetic Change • ∆G = (rg.p * i * ∂g ) • Where: rg.p = Accuracy of Selection • i = Intensity of Selection and • ∂g = Genetic Variability (standard deviation of the BV) rg.p -- Accuracy of Selection • Definition: Correlation between the breeding value of the animal and the phenotypic information used to estimate it • Individual Selection = Picking an animal only on his own performance, Yearling Weight, for example • Accuracy = h (the square root of heritability) • As h2 increases, so does the accuracy of picking the animal from his own phenotype Effect of h2 on Accuracy of Selection h2 .1 .3 .5 .7 rg.p .32 .55 .71 .84 As the heritability of a trait increases, an animal’s own performance (weight or milk production, etc.) gives a more accurate indication of that animal’s breeding value rg.p -- Accuracy of Selection • HOW GOOD ARE WE AT PICKING THE BEST ONES ON AN ADDITIVE GENETIC BASIS? • MORE SPECIFICALLY, HOW ACCURATELY CAN WE RANK A GROUP OF ANIMALS (BULLS, SAY) BASED ON THE GENOTYPE USING PHENOTYPIC MEASUREMENTS TO DO SO • IF THE TRUE ORDER IS 1-2-3-4-5-6-7-8-9-10 • DO WE RANK/ORDER THEM? • 3-1-2-4-6-5-8-9-7-10 (fairly close to the real ranking) OR • 7-3-4-1-6-2-10-8-9-5 (not SO very close) ACCURACY OF SELECTION REPEATED RECORDS Where n = the number of records measured and R = the repeatability of the trait Accuracy of Selection can be increased with multiple records If h2 = 0.1 and R = 0.2 such as for litter size in swine and n = the number of records per animal n rg.p 1 .32 2 .41 3 .46 4 .53 5 .60 Accuracy increases with each additional repeated record, BUT 1. Increases in accuracy with additional records become smaller with each additional record. 2. The additional time required to collect the additional records delays selection decisions. INCREASING ACCURACY OF SELECTION BY REPEATED RECORDS Means of Increasing Accuracy of Selection • Repeated measurements • Comparing animals under more uniform environmental conditions (goal is ↑h2) • Mathematically correct for (that is, adjusting the data) non-genetic effects – (Ex. Age of calf, Age of dam, Season of the year, etc.) • Use of measurements of performance on relatives– Sire, Dam, Siblings, Progeny • Use of measurements of traits that are genetically correlated with economic traits Progeny Testing • Evaluating the genotype of a sire (or a dam) based on the phenotypic performance of their progeny • Especially important for sex-limited traits – Traits expressed in females but not males – Milk production, litter size • Progeny testing is also more important in traits that are lowly heritable Daughters = Progeny SIRE 7HO07615 SOLID-GOLD COLBY-ET USE OF PROGENY TESTING TO INCREASE ACCURACY OF SELECTION Examples of the Increase in Accuracy through Progeny Testing No. of Progeny 10 t = h2/4 h2 = .20 for milk rg .5869 20 .7157 50 .8508 100 .9167 1000 .9902 sire .pprogeny COMPARISON OF THE ACCURACY OF SELECTION FROM PROGENY TESTING TO INDIVIDUAL SELECTION ACCURACY OF SELECTION OR (How good are we at picking the best ones?) Heritability h2 = .10 h2 = .25 h2 = .50 rGP .32 .50 .71 .23 .35 .50 .71 .85 .92 .73 .87 .93 (individual performance) rGP (both parents’ performance) rGP (40 progeny performance) rGP (individual performance + 40 progency) Intensity of Selection • Intensity of Selection is calculated by dividing the Selection Differential by the standard devaition of the trait • i = (Mean of Selected Parents – Overall Mean) Standard Deviation of the Trait Intensity of Selection • • • • • Example: Yearling Weight Selection in Bulls Mean Yearling Weight = 1100 Standard Deviation for Yearling Weight = 70# If the average of the bulls kept for breeding = 1240, then the Selection Differential is 140 • Thus, i = 140/70 = 2 Intensity of Selection • The key concept associated with Intensity of Selection is that it is directly related to the proportion of the animals selected for breeding (to produce the next generation) • As we decrease the proportion selected (pick a smaller % of increasingly superior animals as parents), we increase the intensity of selection Intensity of Selection Proportion Selected As Parents 100 0 50 0.8 15 1.55 5 2.06 1 AI Intensity of Selection 2.66 0.5 2.90 Intensity of Selection • Intensity of Selection can be increased by: – Artificial Insemination as opposed to Natural Service – Superovulation/Embryo Transfer vs. Natural Service – Using more highly fertile species – chickens vs. elephants – Sexed Semen in dairy cattle – if 80% of the calves are heifers, we don’t need to keep all of them “A NEW WRINKLE” • “Time is Money” • HOW FAST WE MAKE GENETIC PROGRESS IS OF MAJOR IMPORTANCE! • THE GENERATION INTERVAL MEASURES HOW LONG EACH GENERATION PERSISTS BEFORE IT IS REPLACED BY THE NEXT GENERATION • It is NOT the same as AGE at PUBERTY!!!! • Horses, for example, are not bred for the first time until long after they have reached puberty ∆G/Year = (rg.p * i * ∂g ) / Generation Interval • Where: – rg.p = Accuracy of prediction of the breeding value – i = Intensity of Selection, inversely proportional to the proportion of the population selected and which will be different for males than females And, ∂g = the genetic variability remaining in the population Generation Interval • Average Age of the Parents of the “Replacement Animals” – the next generation • Turnover Rate – How fast the current generation is replaced with the next generation • Can be very quick, say in Drosophila about 2 weeks • Could be very long, say in Elephants, perhaps 20 years GENERATION INTERVAL IS NOT • Generation Interval is not the same as age at puberty • Generation Interval is almost always much longer than age at puberty as sires and dams can and do continue to produce animals that might be saved as replacement animals throughout their productive lives • Example 20+ year old stallions and mares BEEF CATTLE GENERATION INTERVAL (YEARS) 4.5 - 5 DAIRY CATTLE 4-5 SHEEP 4-4.5 SWINE 2.5 CHICKENS 1.5 SPECIES THOROUGHBRED 9 - 13 HORSES Effect of Progeny Testing on the Generation Interval • PROGENY TESTING WILL LENGTHEN THE GENERATION INTERVAL!!!!!!!!!! • WHY? YOU MUST WAIT FOR THE PROGENY TO BE BORN, GROW UP AND PRODUCE BEFORE SELECTION CAN OCCUR • DAIRY BULLS ARE BETWEEN FIVE AND SIX YEARS OLD WHEN FIRST PROGENY TESTED • SINCE GI IS IN THE DENOMINATOR OF THE FORMULA FOR ∆G/Year, IT CAN DECREASE IT! Effect of Progeny Testing on Generation Interval and rg.p • = (rg.p * i * ∂g ) / GI • PT INCREASES GI as well as the rg.p • THUS, PT MUST INCREASE rg.p MORE THAN GI! This occurs primarily with sex-limited traits, traits low in heritability and traits that require slaughter to be measured PT MUST INCREASE rg.p MORE THAN GI! • (rg.p * i * ∂g ) / GI • The numerator must increase more than the denominator! • With highly heritable traits, accuracy may already approach 1 (its maximum) or at least be quite high (> .7) For example, at trait with a heritability of .6 √.6 = .77, probably GI will increase more by more than 30% through PT, thus not gaining anything due to the effort of PT Genomic Testing: It changes the picture a lot! • The accuracy of a young Holstein bull based on pedigree information alone was about .35 • The data would include dam’s PTA based on her milk yield and the sire’s PTA based on his daughters, primarily • Nowadays, bulls of one year of age with the combined effects of genomic testing and progeny information have an accuracy of .60 to nearly .80 depending upon the trait! Genomic Goals • Predict young bulls and cows more accurately • Compare actual DNA inherited • Use exact relationship matrix G instead of expected values in A • Trace chromosome segments • Locate genes with large effects Genetic Correlations Between Traits • Genetic Correlation: A measure of the strength (consistency, reliability) of the relationship between breeding values for one trait and breeding values for another trait • The genetic correlation, then, tells us how much of the genetic influence on two traits is common to both: if it is above zero, this suggests that the two traits are influenced by common genes. Genetic correlations measure relationships between traits: • Genetic correlations can be positive or negative and range from -1.0 to 1.0. • Genetic correlations tell us how pairs of traits "covary" or change together. When genetic correlations are close to zero, different sets of genes control each trait and selection for one trait will have little effect on the other. • As genetic correlations become different from 0.0, then more of the same genes affect both traits. • Selection for one trait will increase the other if the genetic correlation is positive and decrease it if the genetic correlation is negative. Causes of Genetic Correlations • PLEIOTROPY – THE SAME GENES CONTROL MORE THAN ONE TRAIT • LINKAGE – THE GENES CONTROLING THE TWO TRAITS ARE LOCATED ON THE SAME CHROMOSOME – LIKELY TO BE ONLY TEMPORARY AS CROSSING OVER WILL ELIMINATE THE CONNECTION Genetic Correlations Among Growth Traits in Beef Cattle Koch et al., 1974 GENE 1 GENE 2 GENE 3 GENE 4 GENE 5 GENE 6 GENE 7 GENE 8 GENE 9 GENE 10 GENE 11 GENE 12 BIRTH WEIGHT YEARLING WEIGHT Environments Can Also Cause Correlations Among Traits • Environments can also cause relationships among traits • Example: Extra feed will cause both weaning weight and yearling weight to increase • This is not genetic and will not influence subsequent generations! Crews, Bennet Cassell, Virginia Tech CORRELATED RESPONSE to Selection • When selection is done on a trait, other traits tend to either change independently, vary in the same direction (positive correlation) or in the opposite direction (negative correlation). • From 0.7 to 1.0-traits will change strongly together; • From 0.35 to 0.7-traits will change somewhat together; • From 0 to 0.35-traits will change almost independently of one other. Correlated Responses to Selection • Correlated responses to selection occur when we select for Trait A and change it. • When we change Trait A, any other trait that is genetically correlated with it will also be expected to change. • Example: As we select for and increase YW in beef cattle, we should expect BW to also increase as the genetic correlation between the traits is about 0.6 Correlated Responses in Dairy Cattle • The genetic correlation between fat % and protein % in dairy cattle is high (~.8) as a result, as we select for increased protein %, fat % would also be expected to increase substantially • The genetic correlation between fat % and milk yield is negative (-0.4), as a result selection for increased milk yield will tend to decrease fat %. Indirect Selection • Selecting on one trait that is easier to change for one reason or another when what we want is an improvement in a trait that is genetically correlated with that trait • Indirect Selection is not commonly practiced in food animal species • One possible example is selection based on scrotal circumference when what we are interested in changing is age at puberty in heifers h2 = .25 Breeding Value hap (Age at Puberty in heifers) (Age at Puberty in heifers) rg (ap-sc) Breeding Value (Scrotal Circumference) Phenotype h2 = .50 hsc Phenotype (Scrotal Circumference) Effectiveness of Indirect Selection • Correlated Response Direct Response Selection on bulls for SC with the intention of reducing AP in heifers Selecting directly on heifers for earlier age at puberty • = hsc * i sc * rg (ap-sc) = (.71)*(1.6)*(.9) _____________________ hap * i ap (.5) * (0.8) • = 2.55; thus indirect selection is more than twice as effective Why is indirect selection likely to be so effective? • The heritability of scrotal circumference is twice as high as that of age at puberty in heifers • The intensity of selection possible for bulls when selecting for scrotal circumference is at least twice as high as that for age at puberty in heifers • The genetic correlation between the two traits is very high – both are really measuring age at puberty Undesirable Effects of Genetic Correlations • Selection for increased yearling weight results in increased birth weights and thus more dystocia • Selection for increased milk production can possibly lead to: – Lower fat % in milk – Lower pregnancy rates – Increased health issues ...
View Full Document

Ask a homework question - tutors are online