Repeatability | Animal Husbandry & Veterinary Science Optional for UPSC PDF Download

Introduction

Repeatability is a concept closely related to heritability, particularly applicable to traits expressed multiple times during an animal's life. This includes characteristics such as milk yield in dairy cows, number farrowed, and litter weight in swine, or weaned weight in lambs or calves for sheep or beef cattle.

Comparison with Heritability

• Repeatability, expressing the consistency of trait expression over time, should be at least as large as heritability in a given population. 
• It might even be larger due to the inclusion of certain permanent environmental influences in the repeatability fraction.

Repeatability Estimates for Expressions Over Time

Trait Expression at Different Life Stages

• Repeatability estimates focus on traits expressed multiple times during an individual's lifetime, such as milk production, fleece weight, etc. 
• There is no segregation or independent assortment of genes, making repeated expressions predictable.

Prediction from Early Expressions

• For instance, the weaning weight of a calf from a first-calf heifer can predict future breeding potential. 
• A significant difference in weaning weight during the first lactation suggests consistent differences in later years.

Fraction of Differences in Future Records

• Another perspective on repeatability is that it represents the fraction of differences between single records of individuals likely to occur in future records. 
• For example, a 16% repeatability for litter size at weaning in swine suggests a predictable average difference in later litters.

Computation of Repeatability

Regression of Future Performance on Past Performance

Repeatability can be computed as the regression of future performance on past performance. This method, along with an analysis of variance and correlation among records or observations of traits, provides insights into trait consistency.

Components of Repeatability

Repeatability is expressed as the sum of the genetic variance (Vg), variance associated with permanent environmental influences (Vpe), and residual variance (Ve).

Understanding Permanent Environmental Influences

Permanent environmental influences, like a cow losing one quarter permanently or the effects of poor feeding and care on young animals, contribute to the variance in trait expression over time.

Applications of Repeatability

Gaining Insights from Repeated Measurements

Repeatability is valuable for understanding the potential gain from repeated measurements and setting upper limits for certain ratios. It also sheds light on the nature of environmental variance in trait expression.

Repetition of Traits: Temporal and Spatial Aspects

Temporal Repetition and Examples

• Repetition of traits occurs through temporal and spatial means. Milk yield and litter size exemplify temporal repetition, where measurements are taken across successive lactations or pregnancies.

Spatial Repetition

• Spatial repetition involves measuring a trait across different individuals within a population. This approach aids in analyzing variance components within and between individuals.

Analyzing Variance Components

Within Individual Component

• The variance within individuals captures temporary environmental differences affecting an individual's successive performances.

Between Individual Component

• This component encompasses both environmental and genetic factors, with the environmental aspect stemming from circumstances permanently influencing individuals.

Prediction of Future Performance

No Genetic Connotation

• Prediction of future performance, while lacking genetic implications, relies on understanding the variance partitioning into permanent and temporary effects, facilitated by repeatability analysis.

Consideration of Population Mean

• Performance deviations from the population mean, past and future, shape predictions. Temporary environmental effects on past performance do not carry forward, causing future performance to regress toward the population mean.

Importance of Population Characteristics

Accurate prediction hinges on understanding population characteristics, especially regarding traits like milk yield in first and second lactation. The repeatability, indicating the correlation between the two performances, guides the accuracy of predictions.

Methodology for Daughter-Dam Comparisons

Even Number of Daughter-Dam Comparisons

• If a bull had an even number of daughter-dam comparisons, all were considered. In cases of an odd number, the mate with a median first record was discarded along with her daughter's information to ensure an equal number of mates for each bull in both high and low groups.

Equalizing Herd Averages

• This approach was designed to eliminate the influence of differences in herd averages or the genetic merit of the sires on the comparisons between high and low groups.

Total Variability between Mate's First Records

Figure 1 illustrates that the difference (X) between the mate's first records was 40.1 kg of fat, representing the total variability between the two groups of cows.

Real Difference in Producing Ability

The difference (Y) between the later records of these groups of cows was 198 kg, representing the real difference in producing ability among these cows if successive records were obtained.

Selection Bias and Environmental Circumstances

Selection Bias in Low Half

By dividing mates into low and high halves, the low half represented cows with lower inherent ability, and due to selection based on record size, poorer than average environmental circumstances were also represented.

Inclusion of New Sample of Herd Environment

Successive records introduced a new sample of herd environment, assumed to be representative for both low and high groups, highlighting only the repeatable or real differences in later records.

Estimating Repeatability

This methodology provides an approximation to the regression of future performance on present performance, serving as an estimate of repeatability. In this case, the repeatability of differences in single lactation fat yields is calculated as 198/40.1.

Analysis of Daughter Lactations

Averages for daughter lactations are displayed in Figure 1. Daughters from the high group of dams averaged 744 kg, while those from the low group averaged 1720 kg.

Consideration of Sire Differences

Each sire was equally represented by daughters in both low and high groups. Sire differences did not contribute to the 6.4 kg difference between the daughters of these two groups.

Adjustment for Genotype Inheritance

Since daughters received only half of the dam's genotype, the 6.4 kg difference represents only half of the expected difference if both sires and dams could have been divided into comparable low and high groups.

Fig: Regression of later records and daughter's records on the dam's selected record. (Dotted line represente mid-point between selected high sind selected low records)