A general method of detecting additive, dominance and epistatic variation for metrical traits. V. Triple test cross analysis of disomically inherited traits expressed in triploid tissues

Abstract

The applicability of the triple test cross design to the genetic analysis of metrical traits that subscribe to disomic inheritance but are expressed in a trisomie state has been investigated both theoretically and experimentally. Theory has shown that the standard sets of triple test cross families (L_1i etc.) do not provide unambiguous tests of the additive, dominance and epistatic effects when reciprocal crosses are analysed separately. Analysis of the backcross families also suffers from similar problems but only in respect of the additive component and the tests of dominance and epistasis are not biased by the parentage of the families. Selfs of the standard families, on the other hand, do not display reciprocal differences (of heritable kind) and therefore provide umambiguous tests of the additive, dominance and epistatic effects, but the dominance component is now detected with reduced reliability as the level of heterozygosity is halved due to selfing. Theory further shows that biases of the various tests are eliminated rather easily by including the reciprocal families in the analysis. This is confirmed to a large extent by the analysis of amylose content in rice which also reveals that it is controlled by genes that display both interallelic (additive and dominance) and nonallelic interactions. Furthermore, dominance is shown to be partial but the dominance ratio seems to be high for both the h_a1 and h_a2 types of non-additive effects.