Gafsi N, Martin O, Bidan F, Grimard B, Puillet L
Université Paris-Saclay, INRAE, AgroParisTech, UMR Modélisation Systémique Appliquée aux Ruminants, 91120, Palaiseau, France; Institut de l'Elevage, F-75595, Paris, France.
Université Paris-Saclay, INRAE, AgroParisTech, UMR Modélisation Systémique Appliquée aux Ruminants, 91120, Palaiseau, France.
Theriogenology. 2025 Mar 1;234:83-91. doi: 10.1016/j.theriogenology.2024.12.005. Epub 2024 Dec 4.
In the context of agroecological transition, breeding females with robust reproductive performance, leading to prolonged lactation sequences, is valuable for farmers. This study aimed to explore the relationship between artificial insemination (AI) success and phenotypic lactation curves that serve as proxies for key biological functions in Alpine and Saanen goats. Using data from two French experimental farms (1996-2021), the study analyzed time series data on milk yield (MY), body weight (BW), and sternal body condition score (BCS_S). These data were modeled at the lactation scale to characterize dynamic profiles and create clusters. Each phenotypic lactation curve was evaluated with three levels of detail: cluster membership, synthetic indicators, and model parameters. To investigate AI success, three datasets were used: 638 lactations with complete MY, BW, and BCS_S data; 1359 lactations with MY and BW; and separate sets with 1731 MY and 795 BCS_S records. A mixed logistic regression model (year as a random effect) assessed the relationship between AI success and phenotypic lactation curve characteristics. Results showed that for primiparous goats, AI success was influenced by MY clusters (p < 0.05), while in multiparous goats, MY and BCS_S clusters did not influence AI success. However, indicators such as persistency (p < 0.001) and BW repletion speed (p < 0.001) were significant. Overall, the lactation curve shape was more important to AI success than milk production level, offering insights for enhancing reproductive performance in dairy goats.
在农业生态转型的背景下，培育具有强大繁殖性能、能延长泌乳序列的母羊对养殖户来说很有价值。本研究旨在探索人工授精（AI）成功率与表型泌乳曲线之间的关系，这些曲线可作为高山山羊和萨能山羊关键生物学功能的代表。该研究利用来自两个法国实验农场（1996 - 2021年）的数据，分析了产奶量（MY）、体重（BW）和胸骨体况评分（BCS_S）的时间序列数据。这些数据在泌乳尺度上进行建模，以表征动态特征并创建聚类。每条表型泌乳曲线都从三个详细程度进行评估：聚类成员、综合指标和模型参数。为了研究人工授精的成功率，使用了三个数据集：638次泌乳的完整MY、BW和BCS_S数据；1359次泌乳的MY和BW数据；以及分别包含1731条MY记录和795条BCS_S记录的数据集。一个混合逻辑回归模型（以年份作为随机效应）评估了人工授精成功率与表型泌乳曲线特征之间的关系。结果表明，对于初产山羊，人工授精成功率受MY聚类的影响（p < 0.05），而在经产山羊中，MY和BCS_S聚类并不影响人工授精成功率。然而，诸如持续性（p < 0.001）和BW恢复速度（p < 0.001）等指标具有显著意义。总体而言，泌乳曲线形状对人工授精成功率的影响比产奶水平更为重要，这为提高奶山羊的繁殖性能提供了见解。
