Genetics of quantitative traits in Arabidopsis thaliana.

The genetic control of 22 quantitative traits, including developmental rates and sizes, was examined in generations of Arabidopsis thaliana derived from the cross between the ecotypes, Columbia (Col) and Landsberg erecta (Ler). The data were obtained from three sets of families raised in the same trial: the 16 basic generations, that is, parents, F(1)'s, F(2)'s, backcrosses, recombinant inbred lines (RILs) and a triple test cross (TTC), the latter produced by crossing the RILs to Col, Ler and their F(1). The data were analysed by two approaches. The first (approach A) involved traditional generation mean and variance component analysis and the second (B), based around the RILs and TTC families, involved marker-based QTL analysis. From (A), genetic differences between Col and Ler were detected for all traits with moderate heritabilities. Height at flowering was the only trait to show heterosis. Dominance was partial to complete for all height traits, and there was no overdominance but there was strong evidence for directional dominance. For most other traits, dominance was ambidirectional and incomplete, with average dominance ratios of around 80%. Epistasis, particularly of the duplicate type that opposes dominance, was a common feature of all traits. The presence of epistasis must imply multiple QTL for all traits. The QTL analysis located 38 significant effects in four regions of chromosomes I, II, IV and V, but not III. QTL affecting rosette size and leaf number were identified in all four regions, with days to maturity on chromosomes IV and V. The only QTL for height was located at the expected position of the erecta gene (chromosome II; 50 cM), but the additive and dominance effects of this single QTL did not adequately explain the generation means. The possible involvement of other interacting height QTL is discussed.

The genetical basis of hybrid vigour in a highly heterotic cross of Nicotiana tabacum.

The genetical control of F1 heterosis, observed in a cross of desirable Nicotiana tabacum varieties, was investigated by analysing the data of the basic generations, triple test cross-families and random samples of doubled haploids (DH) and single-seed descent (SSD) lines. Analyses of the first-degree statistics revealed a complex control underlying the genetic variation, including the presence of epistasis, linkage, maternal effects and their interactions, in addition to the additive and dominance effects of the genes segregating in the cross. These analyses identified gene dispersion, directional dominance, and duplicate epistasis, as the main causes of heterosis. The triple test-cross analysis also confirmed the presence of non-allelic interactions and indicated that the dominance ratio, although inflated by epistasis, is consistently partial for all the traits. The extent of transgression in the recombinant inbred lines finally established unequivocally that, as in numerous other crosses, gene dispersion and unidirectional, but partial, dominance are the true causes of heterosis in this cross too.

Estimation of the true additive genetic variance sigma ki = 1 d2i in the presence of linkage disequilibrium.

Accurate estimates of the true additive genetic variance (sigma ki = 1 d2i) of a cross between two pure breeding varieties can be obtained from the additive genetic components of the first three ranks (D1, D2 and D3) when the latter are biased by the presence of linkage. Additive genetic variances of the lower ranks are directly equatable with sigma ki = 1 d2i because they incur minimal bias even when the predominating linkages are strong. More precise estimates of sigma ki = 1 d2i are however obtainable from the asymptotic regression analysis or a weighted least squares analysis. Estimates of sigma ki = 1 d2i when obtained from 784 hierarchically derived F7 families of the V2 X V12 cross of Nicotiana rustica were observed to be considerably larger than the additive genetic variance displayed by the F13 inbreds of the same cross for all the characters that showed significant excess of repulsion linkages. These results lend support to our commonly held view that the prediction procedures generally underestimate the probability of successful recovery of superior recombinant inbreds.

Predicting the properties of first cycle inbreds and second cycle hybrids extractable from two, three and four parent crosses.

Standard biometrical genetical models of Mather and Jinks (1982) when made applicable to the means and variances of the 55 early generations produced by crossing four parents and six F1's in all possible combinations provide estimates of genetic parameters that can be used to predict the distributive properties of the first cycle inbreds and second cycle hybrids which could be extracted from any of these generations. Thus we can predict the inbreeding and outbreeding potentials of each generation in the early stages of a breeding programme and formulate the best breeding strategy for harnessing the full genetic potential of the breeding material and choosing the best end product. The 55 generations provide reliable estimates of the predictors and therefore should be used whenever possible. Simpler experiments consisting of the basic generations of the six single crosses, however, are sufficient for obtaining estimates of the predictors of the inbreds and can be used to predict their properties when information about the second cycle hybrids is not required or the remaining generations are not available. Replacement of the F2, backcross, three way cross and double cross generations with their randomly mated progenies is expected to improve the accuracy of predictions in the presence of a linkage disequilibrium. However, the gains made may not be justified against the costs of the additional breeding and the delay in making the predictions. Extensive experimental testing of the theory of these procedures must await the completion of the current field experiments. In the meantime the limited tests conducted on the material extracted from the Nicotiana rustica varieties V1, V2, V5 and V12 have confirmed the familiar conclusions that the V1 X V5 and V2 X V12 differ in their inbreeding and outcrossing potentials.

Repeatability of stability estimators for downy mildew incidence in pearl millet.

Repeatability of mean downy mildew (Sclerospora graminicola (Sacc.) Schroet.) incidence, regression coefficients and deviation mean squares were investigated for 25 pearl millet (Pennisetum typhoides (Burm.) Stapf. & Hubb.) genotypes in 20 environments by correlating arrays of these stability parameters over subsets of the 20 environments arranged according to the year-wise, random, stratified and extreme methods of environmental division. Correlation coefficients between arrays of mean downy mildew incidence from different pairs of subsets ranged from 0.57 to 0.98 and those of deviation mean squares from 0.58 to 0.96 indicating good repeatability of these parameters. Arrays of regression coefficients from different subsets, on the other hand, showed correlation coefficients that ranged from -0.58 to 0.96. Apparently, the regression index of stability was not repeatable for the genotypes and environments studied. Therefore, in order to identify a widely adapted genotype, testing is required to be carried out over a wider range of environments.

Predicting the properties of second cycle hybrids produced by intercrossing random samples of recombinant inbred lines.

Distributive properties of the second cycle hybrids that are produced by inter-crossing the recombinant inbreds extractable from the F2 of a cross between two pure breeding lines can be predicted from the early generations of the original cross. Hence the frequency of such hybrids that will outperform the extreme recombinant inbreds or the original F1 can be predicted. Basic generations and triple test cross families provide the most reliable estimates of the predictors and therefore should be used whenever possible although, in the presence of linkage, randomly mated F2's may give improved predictions. Simpler experiments consisting of the parental varieties and their F1 and F2 generations, however, provide all the information that is likely to be necessary for most practical purposes. The predictive power of the new approach is demonstrated on material extracted from the cross of varieties 1 and 5 of Nicotiana rustica. The predictors were estimated from the means and variances of V1, V5, F1 and F2 raised in six environments between 1973 and 1983. The predicted frequencies of second cycle F1's which outperform the extreme recombinant inbred lines derived from this cross are compared with those observed among 190 second cycle hybrids in a diallel between 20 recombinant inbreds derived from the same cross.