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

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.

鉴于奶牛饲料效率（feed efficiency）数据集较为匮乏，采用模拟数据集是探索不同环境下育种策略相关性的可行途径。本数据集依托一套机制模型（mechanistic model），可模拟不同营养环境下奶牛全生命周期的表型轨迹（phenotypic trajectories）（Puillet等，2016；Puillet等，2021）。该模型假定有4个输入参数受遗传调控（genetic control），且用户可自定义营养环境参数，以适配现有及未来的各类场景。当前亟需探究模型中预设的遗传输入参数（genetic parameters）对模拟结果的影响，以及遗传参数与预设环境间的潜在互作效应。通过本随附程序生成的模拟数据集构成了敏感性分析（sensitivity analysis）的基础。研究针对具有系谱结构（pedigree structure）的奶牛群体，模拟了产奶量（milk production）与饲料效率性状。通过调整输入遗传参数，即遗传力（heritability）与表型变异系数（phenotypic coefficient of variation），以及营养环境（饲料供给量），可构建不同的模拟场景。本研究确定了模型中需作为输入的遗传参数，以模拟具有真实均值及遗传（共）方差（genetic (co)variances）的奶牛群体产奶量与饲料效率性状。在所有输入参数中，营养环境对模拟性状间的遗传相关（genetic correlations）影响最大。该模拟工具有望为不同营养环境下奶牛饲料效率的选育策略提供基准测试（benchmark）平台。本数据集包含编译后的可执行文件（executable）及一套脚本，可用于启动模拟、估算遗传参数并开展统计分析。