Skip to main content
ARS Home » Southeast Area » Raleigh, North Carolina » Soybean and Nitrogen Fixation Research » Research » Publications at this Location » Publication #295768
Title: Genetic and phenotypic correlations of quantitative traits in two long-term randomly mated soybean populations

Author
item RECKER, JILL - North Carolina State University
item BURTON, JOSEPH - North Carolina State University
item CARDINAL, ANDREA - North Carolina State University
item Miranda, Lilian

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/15/2013
Publication Date: 3/21/2014
Citation: Recker, J., Burton, J., Cardinal, A., Miranda, L.M. 2014. Genetic and phenotypic correlations of quantitative traits in two long-term randomly mated soybean populations. Crop Science. Vol. 54, No. 3, pp. 939-943.

Interpretive Summary: Random mating aided by genetic male sterility facilitates the development of populations with a broad genetic base. Populations RSII and RSIII were developed using two diverse sets of parents and were random mated without selection for 26 generations. This long term random mating offers a unique opportunity to study the effects of genetic recombination and natural selection on genetic parameters. Genotypic and phenotypic correlations were estimated in both populations to determine the effect of long term random mating. Traits evaluated included: flowering date, maturity date, plant height, lodging score, seed yield, seed weight, protein concentration and oil concentration. The results showed that large genetic correlations were significantly reduced for most traits except for the correlation between protein and oil concentration. It was concluded that pleiotropic effects and not genetic linkage are the most likely cause for the negative association between these two traits.

Technical Abstract: The genetic effects of long term random mating and natural selection aided by genetic male sterility (gms) were evaluated in two soybean [Glycine max (L.) Merr.] populations designated: RSII and RSIII. These populations were evaluated in the field at three locations each with two replications. Genotypic and phenotypic correlations were estimated to determine the effects of 26 generations of random mating. Data was collected on flowering date (R2), maturity date (R8), plant height, lodging, yield, seed weight, protein concentration and oil concentration. The genotypic correlation between yield and protein was not significant and small (less than |0.2|) in both populations however, negative in RSII. The genotypic correlation between yield and oil was not significant but positive in both populations. In contrast, for both populations, the genotypic correlations between protein and oil were significantly negative and large (greater than |0.5|). As these populations were randomly mated for many generations, linkage phase disequilibrium (LD) is likely close to zero; therefore it was concluded that pleiotropic effects, likely caused by interrelated physiological processes, maintain the negative association between oil and protein. The genetic correlations between yield and protein/oil in these populations however, do suggest that with the right breeding scheme, the negative associations can be reduced and allow for improvement in more than one trait simultaneously.