QTL Mapping and Interaction Analysis of Seed Weight per Plant in Soybean among Different Environments

The objectives of this study were to map QTLs, which stably controlled seed weight per plant in soybean and analyze their epistatic effects and QTL environment (QE) interactions. A set of F 2:14-F 2:18 RIL populations with 147 lines were planted in two sites in Heilongjiang Province, China from 2006 to 2010, and the QTLs were deduced using CIM and MIM models simultaneously. Seventeen QTLs for seed weight per plant were located in D1a, B1, B2, C2, F, G, and A1 linkage groups, which explained 6.0–47.9% of the phenotypic variation. Three QTLs could be detected by both methods, with phenotypic contributions ranging from 6.3% to 38.3%. Four QTLs could be found in at least two years, which explained the variation in seed weight per plant by 8.1–47.9%. Data from seven environments were used to detect QE interaction effects and epistatic effects using QTLMapper. One QE QTL and four pairs of epistatic QTLs were detected, but their additive effects and phenotypic contributions were small. These results indicated that both major and minor QTLs with epistatic and QE interaction effects should be considered in soybean breeding aiming at seed weight improvement.