Related Experiment Video
Updated: Jun 3, 2026
04:20
Limited Bedding and Nesting as a Model for Early-Life Adversity in Mice
Published on: July 12, 2024
Linear versus nonlinear offspring-parent regression in unselected random-bred mice
1Institute of Cell, Animal and Population Biology, University of Edinburgh, Edinburgh, Scotland.
Summary
This study examined offspring-parent body weight regression in mice, finding slight nonlinearity in offspring-sire relationships but not offspring-dam relationships. Ignoring this minor nonlinearity generally leads to small errors in predicting selection response.
Area of Science:
- Animal genetics
- Quantitative genetics
- Animal breeding
Background:
- Understanding the linearity of regression is crucial for accurate estimation of heritability and prediction of selection response in animal populations.
- Nonlinear relationships can complicate genetic analyses and potentially bias breeding value predictions.
Purpose of the Study:
- To assess the linearity of the regression of offspring 6-week body weight on parental body weight in unselected mice.
- To determine the impact of potential nonlinearities on heritability estimates and the prediction of selection response.
Main Methods:
- Utilized data from 1099 offspring (family mean)-parent pairs across 20 generations of random-bred mice.
- Calculated regression and heritability estimates using single-parent, mid-parent, and simultaneous two-parent regressions.
- Investigated linear and nonlinear relationships between offspring and parental body weight.
Main Results:
- Offspring-sire regressions exhibited significant, though minor, nonlinearity.
- Offspring-dam regressions did not significantly deviate from linearity.
- No significant contribution from sire-dam cross-product terms was detected.
- Data approximated a normal distribution.
Conclusions:
- While a slight nonlinearity exists in offspring-sire regressions for body weight, it has minimal impact on the prediction of selection response when ignored.
- The regression of offspring on dams is effectively linear for this trait.
- These findings support the use of linear models for predicting selection response in similar populations, with minor adjustments for sire-related nonlinearity if high precision is required.
