Unité Modèles pour l'Ecotoxicologie et la Toxicologie (METO), Ineris Parc Technologique Alata, BP2, 60550 Verneuil-en-Halatte, France.
Math Biosci. 2013 Aug;244(2):148-53. doi: 10.1016/j.mbs.2013.05.001. Epub 2013 May 16.
A mathematical model to distinguish mature female and male three-spined sticklebacks Gasterosteus aculeatus L. 1758 is proposed. This method is based on sexual dimorphism in the head morphology. The discrimination was established on five distances of interest on the head, adjusted by the standard length of fish. The parameters were estimated based on a training set composed of 102 fish with an equilibrium sex ratio and validated on a test set composed of 69 fish. The model estimates the relationship between the percentage of fish that can be sexed with our model and the percentage of fish correctly sexed. For instance, to reach 1% of error in the sex determination, only 53% of the fish should be considered, whereas to reach 5% of error, 90% of the fish can be used. Compared to other available methods to sex G. aculeatus, the model is non invasive, not expensive, rapid, replicable, and can be calibrated outside of the breeding period.
提出了一种区分成熟雌性和雄性三刺鱼(Gasterosteus aculeatus L. 1758)的数学模型。该方法基于头部形态的性二型性。通过对头部的五个感兴趣距离进行调整,以鱼的标准长度进行区分。基于具有平衡性别比例的 102 条鱼的训练集估计参数,并使用由 69 条鱼组成的测试集进行验证。该模型估计了我们的模型可以对其进行性别鉴定的鱼的百分比与正确鉴定的鱼的百分比之间的关系。例如,要将性别鉴定的错误率降低到 1%,只需要考虑 53%的鱼,而要将错误率降低到 5%,则可以使用 90%的鱼。与其他可用于鉴定 G. aculeatus 的方法相比,该模型是非侵入性的、经济实惠的、快速的、可复制的,并且可以在繁殖期之外进行校准。