Hybrid development has become an essential component in many major crop species due to the exploitation of heterosis. However, parental combinations that exhibit high heterotic gains can be difficult to obtain if no prior genetic architecture is known. The current research focuses on heterotic gene pool development using genetic distance and multivariate cluster analysis. Seventy-nine inbred accessions were grouped using three different methods including; 1) phenotypic classification based on 20 qualitative and quantitative traits; 2) Sequence related amplified polymorphisms (SRAP) using 29 forward and reverse primer combinations; 3) genotyping-by-sequencing (GBS) using 80,005 single nucleotide polymorphisms. Both genotypic methods (SRAP and GBS) were compared to each other, and hierarchical clustering produced similar results with 68% homology between the two methods. Heterotic cluster accuracy was investigated between the two genotypic methods through hybrid performance. Forty-four hybrid combinations along with parental cultivars were grown at one location in Winnipeg, MB with three replicates in a randomized complete block design. Hybrid yield and heterosis were regressed against the genetic distance from each genotypic method. GBS was a better predictor of hybrids with high yield (R = 0.47, p <.01); however, SRAP was a better predictor for mid-parent and high-parent heterosis at R = 0.53 and R = 0.61, respectively. This research provides experimental evidence that SRAP and GBS heterotic pool definitions have utility in the prediction of high heterotic parental combinations / May 2016
| Identifer | oai:union.ndltd.org:MANITOBA/oai:mspace.lib.umanitoba.ca:1993/31211 |
| Date | 13 April 2016 |
| Creators | Lees, Corey |
| Contributors | Robert W. Duncan (Plant Science), Dilantha Fernando (Plant Science) Anne Worley (Biological Science) Bert Innes (External) |
| Source Sets | University of Manitoba Canada |
| Detected Language | English |
Page generated in 0.0021 seconds