Mapping and analysis of quantitative trait loci in lycopersicon (Tomato) with the aid of genetic markers using approximate maximum likelihood methods

Abstract

1691 F-2 progeny of a cross between Lycopersicon esculentum and L pimpinellifolium grown under field conditions were scored for 18 quantitative traits of economic interest and 10 segregating genetic markers. Each parental strain was homozygous for one allele of each marker. Four of the markers were electrophoretic, and six were morphological. Three pairs of the genetic markers were linked. An algorithm described previously based on maximum likelihood technique was used to estimate the parameters of loci affecting the quantitative traits linked to the genetic markers and the recombination distance between quantitative trait loci and marker loci. The parameters of quantitative trait loci linked to two genetic markers were also estimated by solving for gene effect and recombination frequency from the independent equations derived from each marker. In general there was close correspondence between estimates obtained from the two methods. Except for cases where highest likelihood was obtained at complete linkage, results of the approximate maximum likelihood technique were within the parameter space, i.e. recombination frequencies between zero and 0·5 and positive variance estimates. Unreasonable results were obtained when the assumptions of the method were violated. These results support those presented previously based on simulated results and a positive control indicating that, for samples of this size, accurate estimates are derived by the maximum likelihood technique. A genetic map of quantitative trait loci is presented. © The Genetical Society of Great Britain.