Lay Summary In multiparous species, such as pigs, sheep, rabbits, or mink, measuring birth weight as total litter weight is less laborious than individual measurements. As the trait summarizes the characteristics of a group of animals, the genetic variance of litter weight depends on litter size. In the past, models have been applied that cannot sufficiently account for that feature. As a remedy, we propose a random regression model, allowing for maternal and direct genetic effects, as known from analyses of individual birth weights. Adaptation to variable litter sizes is obtained by multiplying average parental genetic effects by litter size. As a use case, litter weight data from four diverse mouse lines were analyzed, and the results were compared. In conclusion, the proposed random regression model offers new opportunities for a sound and cost-effective utilization of litter weight data in farm animals.
Jahnel et al. (Fri,) studied this question.