Search

Background: Soybean is an important oil and cash crop, soybean protein is the main source of protein for human consumption. Soybean oil is also major edible oil that is rich in unsaturated fatty acids and is more conducive to human long-term consumption than animal oil. An important research direction of current soybean breeding is improvement of the protein and fat content using marker assisted selection and many reports of associated quantitative trait locus (QTL) mapping have been documented. Until to now, a total of 241 QTLs associated with protein content and 315 related to fat content have been collected on the SoyBase website. Methods: In this study, 236 F2 generation plants and a derivative group were constructed by using Jiyu50 and Jinong18, obtained from Jilin Province. Combining three years of phenotypic and molecular detection data, using the ICIM method, one-dimensional scanning detected 24 QTLs related to protein and fat content, two-dimensional ICIM analysis of F2:3 families detected 7 pairs of epistatic QTLs associated with fat content. Result: Based on a mixed model for major genes and polygenes, the C model was determined the optimal genetic model for protein and fat content, with genetic rates of multiple genes of 71.15% and 79.15%, respectively. In the detection of high-oil molecular markers, the detection coincidence degree was Sat_238 > Satt 100> Satt 150 > Satt 636> Sat_287 = Sat_342, while this was Satt150 = Sat_342> Satt100 > Satt 636> Sat_287> Sat_238 for high-protein markers. The QTL localization results showed that 9 microeffect QTLs related to protein content and 7 microeffect QTLs related to fat content were detected. Isolation analysis results were generally similar to those of QTL mapping. One stable QTL associated with protein content and two stable QTLs associated with fat content were identified as being of some application value in soybean molecular marker-assisted breeding. [ABSTRACT FROM AUTHOR]