Sorghum varieties suitable for grain production at temperate latitudes show dwarfism and photoperiod insensitivity, both of which are controlled by a small number of loci with large effects. We studied the genetic control of plant height and flowering time in five sorghum families (A-E), each derived from a cross between a tropical line and a partially isogenic line carrying introgressions derived from a common, temperate-adapted donor. A total of 724 F2:3 lines were phenotyped in temperate and tropical environments for plant height and flowering time and scored at 9139 SNPs using genotyping-by-sequencing. Biparental mapping was compared with multiparental mapping in different subsets of families (AB, ABC, ABCD, and ABCDE) using both a GWAS approach, which fit each QTL as a single effect across all families, and using a joint linkage approach, which fit QTL effects as nested within families. GWAS using all families (ABCDE) performed best at the cloned Dw3 locus, whereas joint linkage using all families performed best at the cloned Ma1 locus. Both multiparental approaches yielded apparently synthetic associations due to genetic heterogeneity and were highly dependent on the subset of families used. Comparison of all mapping approaches suggests that a GA2-oxidase underlies Dw1, and that a mir172a gene underlies a Dw1-linked flowering time QTL.

Soybean [Glycine max (L.) Merr.] breeding in the United States currently relies on a narrow genetic base. Glycine tomentella Hayata (2n = 78), native to Australia, is a perennial relative in the tertiary gene pool of soybean. No effort has been devoted to using this species to increase seed yield. The objectives of this research were (i) to identify high-yielding lines derived from backcrosses between the soybean cultivar Dwight (2n = 40) and G. tomentella PI 441001 (2n = 78) and to compare their agronomic performance with that of the recurrent parent, and (ii) to determine associations between G. tomentella introgressions and agronomic traits. PI 441001 was crossed to Dwight and immature seed rescue was used to produce a sterile F1 plant. Amphidiploid plants (2n = 118) were produced by treating the F1 hybrid with colchicine. Amphidiploid plants were backcrossed to Dwight and a series of backcrosses were made to obtain lines with 2n = 40 chromosomes, which were self-fertile and genetically stable. Preliminary yield testing was used to select 180 lines in maturity groups II, III, and IV that had yields greater than or less than that of the recurrent parent. Yield data collected in two replication tests at six to eight locations in 2013 and 2014 identified experimental lines in all three maturity groups that were higher yielding than the recurrent parent. All experimental and parental lines were genotyped, revealing an average of 1% G. tomentella introgression. Significant associations were detected between G. tomentella introgressions and time of flowering, height, lodging, and yield.