221-21 Comparison of Phenotypic Vs. Marker-Assisted Selection Methods for Soybean Isoflavone Genistein.



Tuesday, October 18, 2011
Henry Gonzalez Convention Center, Hall C, Street Level

Christopher Smallwood, Catherine Nyinyi, Dennis West, Carl Sams, Dean Kopsell and Vincent Pantalone, Plant Sciences, University of Tennessee, Knoxville, TN
Soybean isoflavones have gained considerable interest in recent years as a potential benefit to human health.  Among the soybean isoflavones, genistein occurs in the greatest abundance and is associated with the most health benefits.  Using molecular markers to select for increased genistein could potentially be a useful method.  However, for a complex trait like genistein, it is unclear that MAS would be as successful as phenotypic selections.  In order to evaluate the effectiveness of these selection methods, both were tested on a population of 274 RILs derived from the parental lines Essex and Williams 82.  Phenotypic data for genistein was collected in 2009 in the F5:8 generation using near infrared spectroscopy.  Genotyping was done with 1,536 SNPs, of which 480 were polymorphic. Six QTL for genistein were identified on chromosomes 5, 6, 7, 9, 13b, and 19 with composite interval mapping.  High and low tail selections at 2.5% intensity were made from the phenotypic data in each of four relative maturity groups (III late, IV early, IV late, and V early).  The RILs displaying high and low QTL were selected for the MAS tests in all but the III late MG. These selections were grown in 2010 in an RCBD replicated three times in three different environments in Tennessee.  Statistical analysis with PROC MIXED (SAS, 2008) showed that the mean high selections were greater than the mean low selections (P<0.05) for each test except for the V late MAS test, in which this trend was reversed.  The mean phenotypic selections were not different from the mean QTL selections within each maturity group (P>0.05), except for the V late tests (P<0.05).  While MAS showed some improvement in genistein content, making selections based on phenotypic data appears to be a more effective method based on the results from this study.
See more from this Division: C01 Crop Breeding & Genetics
See more from this Session: Plant Breeding and Genetics Graduate Poster Competition