*2.3. Linkage Mapping and Statistical Analysis*

Of 893 SNPs segregating in the RIL population, 128 high quality SNPs were selected for the linkage map construction after excluding SNPs with missing rate over 10%, significant segregation distortion (Chi-square test *p*–value < 0.05), or overlapping genetic positions. The composite interval mapping was implemented using QTL IciMapping version 4.1 [21] with the threshold LOD (logarithm of the odds) score of 3.0. The same program was also used to estimate the additive e ffects and the phenotypic variation explained by each QTL at the peak LOD. The mean comparisons of DH between the RILs with di fferent allelic combinations of the identified QTLs were conducted by the Duncan's multiple range test using SAS version 9.4 (Cary, NC, USA). Three-way factorial ANOVAs were conducted to study the main e ffects of the three DH QTLs and their two-way and three-way interactions using SAS version 9.4.

### *2.4. Sequencing and Marker Analysis for Hd16, Hd1, and Ghd7*

The coding regions of *Hd16*, *Hd1*, and *Ghd7*, the candidate genes underlying the three DH QTLs from the Koshihikari × Baegilmi RIL population, were sequenced to search polymorphisms between Koshihikari and Baegilmi. The tenth exon of *Hd16* was amplified by polymerase chain reaction (PCR) using the primers and conditions described in Supplementary Table S1. The two exons of *Hd1* and the two exons of *Ghd7* were amplified by PCR as described in [22]. Sequencing of the PCR products was performed by Macrogen (Daejeon, Korea). To search polymorphism in the promoter region of *Ghd7*, the 2 kb upstream region of *Ghd7* from Koshihikari and Baegilmi was sequenced using the customized sequencing service at Macrogen (Daejeon, Korea). Molecular markers to di fferentiate the three main polymorphisms in *Hd16*, *Hd1*, and *Ghd7* between Koshihikari and Baegilmi were developed (Supplementary Table S1) and used to screen 295 Korean rice cultivars released in 1979–2017 by NICS, Rural Development Administration.
