罗氏沼虾高世代育种群体收获体重加性和显性遗传效应

精确地估计加性和显性遗传效应,可以提高选择准确度和加速遗传进展。本研究构建了343个罗氏沼虾(Macrobrachium rosenbergii)G7~G9育种群体全同胞家系(半同胞家系244个),测定了29523尾个体的收获体重。基于单性状动物模型,利用平均信息约束最大似然法(average information restricted maximum likelihood method,AIREML)估计了G7、G8、G9和G8+G9 4个数据集收获体重的方差组分。分析时采用了两种模型:(1)加性遗传效应模型,包含加性遗传效应和共同环境效应(A+C);(2)加显性遗传效应模型,进一步包括显性遗传效应(A+D+C)。结果表明,在A+C模型下,估计得到的4个数据集收获体重的遗传力范围在0.046~0.082,为低遗传力水平(h~2≤0.15)。在A+D+C模型下,估计得到的收获体重遗传力范围在0.063~0.096,显性方差组分比率范围为0.027~0.571。模型中包括显性遗传效应后,G8数据集收获体重遗传力的估计值变大,其余3个数据集的估计值变小。罗氏沼虾育种群体收获体重遗传力较低,表明需要引进性能优良的野生或改良群体,增加育种群体的遗传变异丰富度;4个数据集显性遗传方差比率值变化较大,表明需要新的算法并利用更多世代数据提高其估计值的准确性。

Estimation of additive and dominant genetic effects for harvest body weight in advanced generations of Macrobrachium rosenbergii

The accurate estimation of additive and dominant genetic effects is fundamental to improving the accuracy of selective breeding and accelerating genetic gains.This study harvested 29523 Macrobrachium rosenbergii individuals from 343 full-sib families (244 half-sib families) and examined the G7,G8,and G9 generations.The variance components of harvest body weight for four datasets (G7,G8,G9,and G8+G9) were estimated using average information restricted maximum likelihood.Two single-trait animal models were used for the analysis:(1)an additive genetic model comprising additive genetic effects plus common environmental effects (A+C),and (2)an additive-dominant model that includes dominant genetic effects (D) (A+D+C).For the A+C model,heritability estimates of harvest body weight for the four datasets were all low (h2≤0.15),ranging from 0.046 to 0.082.For the A+D+C model,heritability ranged 0.063-0.096,and the ratio of dominant genetic variance to phenotypic variance spanned 0.027 to 0.571.Harvest-body-weight heritability decreased in three datasets (G7,G9,and G8+G9),while increasing in G8.Low heritability estimates indicate that wild or improved populations with strong production performance must be introduced and integrated with the nucleus breeding population.Additionally,large between-dataset differences in the ratio of dominant genetic variance to phenotypic variance suggest that the accuracy of dominance variance estimates should be improved with new algorithms including more generations.