基于AMMI模型分析大菱鲆选育家系基因型与环境互作效应

Analysis of genotype-environmental interaction of turbot(Scophthalmus maximus) based on AMMI model

为选育具有普适性的优良家系和适于某一特定环境的优良大菱鲆家系,实验以大菱鲆选育F2的10个优良家系为材料,利用随机区组设计,将其推广到5个不同的实验点进行养殖实验,采用主效可加互作可乘模型(AMMI)分析方法,分析大菱鲆选育家系基因型与环境互作效应。结果显示,家系和环境互作效应(G×E)的平方和占总平方和的10.82%,达到极显著水平(P<0.01),其对产量差异的影响大于家系基因型间对产量差异的影响,为家系效应的1.44倍;综合双标图AMMI模型分析和稳定性参数分析,家系G5和G8的产量较高,分别为844.796和868.888 g,稳定性参数也最小,分别为1.154 975和2.668 016,属于高产、稳产、广适应性较好的家系,其基因型适于新品系(或新品种)选育;通过双标图分析,5个实验点可被分为3组,E3为一组,E5为一组,E1、E2和E4为一组。家系G6的产量为851.768 g,在环境E3条件下最高产,适于在E3条件下推广;家系G10的产量为911.664 g,在环境E5条件下最高产,适于在E5条件下推广;而G2的产量为784.764 g,在环境E1、E2和E4条件下最高产,适于在环境E1、E2和E4条件下推广。

Turbot （Scophthalmus mco;imus）, a flatfish of deep water species, is a high-nutrition and eco- nomic-valued species and currently cultured widely in China. In recent years, however, serious germ plasm degeneration has come into being due to the lack of the long-term and effective broodstock management programmes, production of farmed turbot became highly unstable and total output declined gradually. Therefore, large-scale family selection programs recurring to PIT tag have been used to initiate the genetic improvement of turbot since national eleventh five-year plan by the Yellow Sea Fisheries Research Insti- tute, Chinese Academy of Fishery Science. At present, F2 population has been finished. In order to obtain new strains （or new varieties） and the extension of fine families,it was highly important, in the process of selective breeding, to obtain fine families with universality and fine families with exclusiveness. Thus, we carried out genotype-environment interaction analysis. In this paper, ten families from turbot selective breeding F2 were dispatched in randomized block design to five farms in different locations （Liaon- ing, Jiangsu, Fujian, Shandong-1, Shandong-2） representing a wide variety of environmental conditions and data of body weight were collected. AMMI （Additive Main and Multiplicative Interaction） Model was used to analyze data collected. The results showed that GxE interaction for body weight was significantly different （P〈 0.01） and G^E effects on body weight was higher than those of families genotype. The for- mer was equal to 1.44 times of the latter. The yield of family G5 and G8 was comparatively high （844.796 g and 868.888 g） and the stability parameter of the two families was the lowest （1.154 975 and 2.668 016）. Obviously, family G5 and G8 had a high and stable yield and wide adaptability and the two kinds of geno- types were beneficial to selective breeding of new variety （or new strain）. Five test locations were divided into three environment groups, i.e., Group E3, Group E5 and Group E1-E2-E4, at the same time, family G6, G10 and G2 should be dispatched to the environment Groups E3, Es, E1-E2-E4, respectively, based on ＂Which- won- where＂ view of the AMMI biplot （the yield of family G6, G10 and G2：851.768 g, 911.664 g and 784.764 g）. The conclusion of the study can provide theoretic basis for selective breeding of new strains （or new varieties） and the extension of fine families.