Coupling genetic and mechanistic models to benchmark selection strategies for feed efficiency in dairy cows: Sensitivity analysis validating this novel approach

Published: 24 June 2022| Version 2 | DOI: 10.17632/txx9zkbd2g.2
Alban Bouquet, Margot Slagboom, Jørn Thomasen, N. Friggens, Morten Kargo, Laurence Puillet


Due to the scarcity of feed efficiency datasets in dairy cattle, using simulated datasets is an option to explore the relevance of breeding strategies under different environments. This data capitalizes on a mechanistic model that simulates phenotypic trajectories of dairy cows over their lifetime under different nutritional environments (Puillet et al., 2016; Puillet et al., 2021). Four input parameters are assumed to be under genetic control. The nutritional environment can also be tailored by users to describe existing and prospective scenarios. There is a need to explore the influence of genetic parameters assumed as inputs in the model on the simulation outputs, as well as potential interactions between genetic parameters and the assumed environment. The datasets simulated with the programs herewith formed the basis of a sensitivity analysis. Milk production and feed efficiency traits were simulated in populations of cows with pedigree structure. Different scenarios were considered by varying input genetic parameters (heritability and phenotypic coefficient of variation) and the nutritional environment (feed offer). We identified genetic parameters to consider as inputs in the model to simulate milk production and feed efficiency traits of populations of cows with realistic means and genetic (co)variances. The nutritional environment was the input parameter with the highest influence on genetic correlations among simulated traits. This simulation tool is promising to benchmark selection strategies for feed efficiency in dairy cows under various nutritional environments. This data comprises a compiled executable and a suite of scripts to launch the simulations, estimate genetic parameters and carry out statistical analyses.


Steps to reproduce

These scripts must be run under a Linux environment using the SLURM workload manager, R software ( and DMU software v5.2 ( must be installed. The simulation software is stochastic. To ensure reproducibility of datasets, a seed is defined within each replicate to initialize the random number generator. First download the zip file, extract all files from it and read the README.txt. A few parameters in the parent shell scripts have to be updated to adapt to your system (mostly SLURM parameters). Those parameters are described in the README.txt.


Institut National de la Recherche Agronomique, AgroParisTech, Aarhus Universitet Faculty of Technical Sciences


Animal Breeding, Dairy Cattle, Agent-Based Modeling