中国农业科学 ›› 2023, Vol. 56 ›› Issue (6): 1177-1188.doi: 10.3864/j.issn.0578-1752.2023.06.013

• 畜牧·兽医 • 上一篇    下一篇

里岔黑猪和巴里杂交猪母猪100kg体重日龄校正方法的比较

崔登帅1(), 熊三亚1, 郑浩2, 李龙云1, 余乃标1, 黄智勇2, 肖石军1, 郭源梅1()   

  1. 1 江西农业大学省部共建种猪遗传改良与养殖技术国家重点实验室,南昌 330045
    2 江西山下华系种猪养殖有限公司,江西赣州 341900
  • 收稿日期:2021-11-29 接受日期:2022-05-17 出版日期:2023-03-16 发布日期:2023-03-23
  • 联系方式: 崔登帅,E-mail:1305568175@qq.com。
  • 基金资助:
    国家自然科学基金(31972542); 国家自然科学基金(31660303)

Comparing Methods for Correcting Days to 100 kg of Sows in Licha Black Pig and Its Intercross with Berkshire

CUI DengShuai1(), XIONG SanYa1, ZHENG Hao2, LI LongYun1, YU NaiBiao1, HUANG ZhiYong2, XIAO ShiJun1, GUO YuanMei1()   

  1. 1 State Key Laboratory of Pig Genetic Improvement and Production Technology, Jiangxi Agricultural University, Nanchang 330045
    2 Jiangxi Shanxia Huaxi Pig Breeding Company Limited, Ganzhou 341900, Jiangxi
  • Received:2021-11-29 Accepted:2022-05-17 Published:2023-03-16 Online:2023-03-23

摘要:

【目的】比较3种达100 kg体重日龄(D100kg)的校正系数CF的校正效果,为种猪场根据自身实际情况选择合适的D100kg校正系数CF提供参考。【方法】以里岔黑猪群体及其与巴克夏杂交1代(巴里1代)和2代(巴里2代)共1 344头猪只为研究对象,利用其生长肥育期体重和日龄的数据,以始测体重≥25 kg、终测体重在80—125 kg、过始测点和终测点直线的斜率在群体平均斜率±2个标准差之内等标准进行质控,然后按照下面描述的三种方法估计里岔黑猪、巴里1代、巴里2代以及它们混合群体达100 kg校正系数CF,并评估它们的校正效果。一是回归校正系数(CFreg):利用测定期间个体i的体重和日龄数据,拟合体重对日龄的一元回归方程,把其斜率记b1i;然后利用它终测点和原点(日龄和体重均为0)作一条直线,把其斜率记为b2ib1i/b2i的平均数就是CFreg。二是两点校正系数(CF2point):利用个体i初测点和终测点作一条直线,得到斜率b1i,斜率b2iCF2point计算方法与CFreg计算方法相同。三是单点校正系数(CF1point):拟合所有个体的终测体重对终测日龄的一元回归方程,得到斜率b1;第二条直线斜率与CFreg相同;b1 / b2i的平均数就是CF1point。分别将CFregCF2pointCF1point代入校正公式,计算出相应的D100kg.CF.reg、D100kg.2point和D100kg.1point。根据前面拟合的日龄对体重的一元回归,计算个体达100kg的回归日龄(D100kg.reg),并计算它与其他D100kg的相关系数和离差。根据相关系数和离差的大小,评估3种方法的优劣。为比较3种校正系数的群体适用性,用某个群体的CF去校正其他3个群体的D100kg,并根据其校正日龄与该群体D100kg.reg相关系数的高低来评估其群体适用性的优劣。通过分析终测体重和校正误差之间的关系,探究终测体重对D100kg的影响。考虑单点校正误差与终测体重之间存在很强的线性关系,利用终测体重对D100kg.1point进行二次校正,以提高单点校正的准确性。【结果】质控后共1 181头母猪用于后续分析。在同一群体中,CFregCF2point的值相近,它们极显著大于CF1point。D100kg.CF.reg和D100kg.2point与D100kg.reg很接近,相关系数均大于0.98,说明这两种方法的校正效果很好。D100kg.1point与D100kg.reg的差异较大,相关系数仅为0.89,校正效果较差。CFregCF2point的群体适应性很好,均可用于其他群体D100kg的校正。CF1point的群体适应性较差,不宜用于其他群体D100kg的校正。4个群体中,终测体重离目标体重(100 kg)越远,D100kg.CF.reg校正误差越大,D100kg.2point校正误差没有明显的变化。D100kg.1point的误差与终测体重均存在极显著的负相关,r分别为-0.9812、-0.9627、-0.9786和-0.9352。经过二次校正后,CF1point校正效果有显著提升,但与CFreg校正效果仍有较大差距,优先选择CFreg。【结论】CFregCF2point的校正效果和群体适应性都很好,可用于生产实践。单点校正系数(CF1point)的校正效果不理想,不宜用于生产实践。在校正D100kg时,可以根据猪只的数据量,采用合适的方法,以提高估计准确性,降低估计误差。

关键词: 里岔黑猪, 达100 kg体重日龄, 校正系数CF, 回归日龄

Abstract:

【Objective】 The aims of this study were to compare the powers of three correction coefficients for days to 100 kg (D100kg), so as to provide the reference for pig breeding farms to select the appropriate correction coefficient for D100kg according to the data they have. 【Method】 A total of 1 344 pigs were used in this study, including Licha black pig, F1 (BL1) and F2 (BL2) from its intercross with Berkshire boars. Body weight and age in days for each pig were recorded everyday during its growth and fattening period. The data were qualified according to following criteria: the start body weight ≥25kg, the final body weight among 80 - 125 kg, the slope of the regression equation of body weight to age in days within±2 standard deviations of its average and so on. Three methods described as following were used to estimate the correction coefficient of the days to 100 kg (CF) into Licha Black pig, BL1, BL2 and their mixed population. First one was the regressive correction coefficient (CFreg): b1i was the slope of the simple regression equation of body weight to age in days for the ith individual, and b2i was the slop of the straight line through the testing end point of the ith individual and the origin (age and weight were both 0), then the average of b1i/b2i in the population was CFreg. Second one is the two-point correction coefficient (CF2point): b1i was the slope of the straight line through the testing start and end points of the ith individual, and the methods to calculate b2i and CF2point were as same as those of CFreg. The last one was the single-point correction coefficient (CF1point): b1 was the slope of the simple regression equation of body weight to age in days at the testing end point for all pigs, and the methods to calculate b2i and CF1point were as same as those of CFreg. Put CFreg, CF2point and CF1point into the correction equation, respectively, and got D100kg.CF.reg, D100kg.2point and D100kg.1point. The simple regression of age in days to body weight was used to calculate the regressive age in days to 100 kg (D100kg.reg), and the correlation coefficients and deviations between D100kg.reg and other D100kg were calculated. The merits and demerits of the three methods were evaluated according to their correlation coefficients and deviations to D100kg.reg. D100kg of a population was estimated by using the correction coefficients from another population, and its correlation coefficient with D100kg.reg of the population was calculated to evaluate whether the correction coefficient estimated from one population could be used in other populations. The relationship between body weight at the end of performance test (BWend) and correction error was also investigated to explore whether BWend had an effect on D100kg. Taking consideration of the strong linear relationship between BWend and the correction error of CF1point, BWend was used to correct D100kg.1point to improve the correction accuracy of CF1point. 【Result】 A total of 1 181 sows were used for following analyses after quality control. In the same population, the values of CFreg and CF2point were almost the same, and both of which were highly significantly greater than CF1point. D100kg.CF.reg and D100kg.2point of each pig were very close to its D100kg.reg, and all of the correlation coefficients among D100kg.CF.reg, D100kg.2point and D100kg.reg were greater than 0.98, indicating that the correction powers of CFreg and CF2point were very good. The difference between D100kg.1point and D100kg.reg was large, and their correlation coefficient was 0.89, so the correction power was not very good. CFreg and CF2point had good population transplantations, and could be used to correct D100kg of other populations, while CF2point had poor population transplantations, and it was unsuitable to correct D100kg of other populations. With the difference of BWend to the target body weight increasing, the correction error of D100kg.CF.reg was also increasing in the three populations, and the correction error of D100kg.2point was no significant changing. D100kg.1point was significant negative correlation with BWend, the correlation coefficients were -0.9812, -0.9627, -0.9786 and -0.9352 in the four populations, respectively. The correction power of CF1point has been significantly improved after adjusting with BWend, but CFreg was still better than CF1point, and it was preferred. 【Conclusion】 The correction powers and transplantations of correction coefficients CFreg and CF2point were very good, so CFreg and CF2point were suitable to be used to correct D100kg in practice. The correction power of CF1point was not very good, therefore, CF1point was unsuitable to be used in practice. According to the datum amount of a pig, its D100kg was estimated with the appropriate correction coefficient in order to improve the estimation accuracy and reduce the estimation error.

Key words: Licha Black pig, days to 100 kg, correction coefficient CF, regressive days to 100 kg