Body trait profiles in Holstein-Friesians modeled using random regression.

Area of Science:

• Animal Genetics
• Quantitative Genetics
• Dairy Science

Background:

• Understanding genetic and phenotypic changes in body traits during first lactation is crucial for dairy cattle breeding.
• Random regression models (RRM) are effective for analyzing longitudinal data in animal breeding.

Purpose of the Study:

• To compare Legendre polynomial and cubic spline functions within RRMs for modeling body trait development during first lactation.
• To assess the genetic variances and heritability of body traits using these modeling approaches.
• To evaluate the predictive accuracy of each function for body type traits.

Main Methods:

• Utilized random regression models with Legendre polynomial and cubic spline functions.
• Analyzed body trait data from 954 sires' daughters across their first lactation (days 50-250).
• Estimated genetic variances, heritability, and predictive accuracy.

Main Results:

• Both Legendre polynomials and cubic splines yielded similar genetic variances for most body traits.
• Heritability estimates were consistent with previous studies.
• Legendre polynomials showed a slight advantage in modeling accuracy and predictive power.
• Identified significant sire-specific differences in growth trajectories during first lactation.

Conclusions:

• Legendre polynomials offer a slightly superior fit for modeling dynamic body trait changes in dairy cattle during first lactation.
• Sire-derived growth profiles can predict future production and functional performance.
• This approach provides insights into the physical and biological development of cows.