*4.2. DNA Extraction and PCR*

DNA was extracted from fresh rice leaves according to the method described in [66], with minor modifications. PCR (polymerase chain reaction) was conducted, and the products were separated by agarose gel electrophoresis [67]. A set of newly developed insertion/deletion (InDel) markers was selected as the primary markers, as found in the open rice genome sequence library (http: //www.ncbi.nlm.nih.gov). Amplified DNA products were analyzed on 3% agarose gels stained with ethidium bromide and photographed with a UVP system.

#### *4.3. Marker Genotyping*

The germplasm comprising 204 rice germplasms was mined for the presence of 9 grain size and weight related genes (i.e., *qPE9~1*, *GW2*, *SLG7*, *GW5*, *GS3*, *GS7*, *GW8*, *GS5*, and *GS2*) using the 14 InDel markers given in Table 8, along with detailed information of the primer pairs for each marker. All the markers were scored visually as 1 for their presence and 0 for their absence.


**Table 8.** Detailed information for the 14 molecular markers for the 9 grain size and weight related genes.

#### *4.4. Statistical Analyses*

Descriptive statistics and factorial analyses for the trait association of GL, GT, GW, and TGW were conducted using the statistical package SYSTAT, version 13.1. Principal Component Analysis (PCA) for testimation of the phenotypic variability within the germplasm was performed using Minitab software, version 18. Correlation coefficients among the grain size and weight related traits were determined using the computer-based software, IBM-SPSS, version 25.

For allele scoring, the genetic variations, expected heterozygosity, and polymorphism information content (PIC) values of the 14 markers were calculated using binary data in the POWER MARKER software, version 3.25 [68]. The presence of alleles was scored as 1, whereas the absence of alleles was scored as 0 to generate a binary matrix for each genotype. An un-weighted neighbour joining UPGMA tree was built based on the marker data using the DARwin 6 software [69]. Another UPGMA dendrogram (to categorize different groups based on phenotypic data) was constructed using the PAST version 3.25 software. A scatter plot was constructed using the PAST version 3.25 software to disperse the germplasm further into the sub-populations based on the InDel marker data [63,70]. The Principal Coordinate Analysis (PCoA) analysis and Analysis of Molecular Variance (AMOVA) were conducted from the binary data of the InDel markers using GenAlEx version 6.502. The binary data of the genetic markers were constructed according to grain size, which was used to separate the total molecular variance between and within the groups. **Φ**PT was also calculated using GenAlEx v. 6.502.

### **5. Conclusions**

Collectively, it can be concluded that allelic variations of the nine genes are extensively distributed in the studied germplasm. Further, we observed that most of these alleles are significantly associated with variations in one or more of the studied traits related to grain size and grain weight in rice. The results also suggest that several genotypes have similar grain size characteristics and share particular allele and/or allele combinations of the nine key genes examined in this study. The examinations of allelic contributions from different genes in regulating grain size and weight related traits undertaken in the study will further strengthen our understanding of the complex mechanisms involved in rice grain appearance and the grain weight to be utilized in rice breeding programs. The identified genes and their linked InDel markers could be highly informative in pyramided breeding strategies and selecting parental lines for developing rice varieties according to consumer needs.

**Supplementary Materials:** Supplementary materials can be found at http://www.mdpi.com/1422-0067/20/19/ 4824/s1.

**Author Contributions:** S.G., H.G. and Z.H. analyzed genotype data for InDel marker development/ Estimation; Z.H. and R.A.R.K. analyzed all statistical analyses; H.G., J.M. and T.W. conducted phenotype experiment; G.L., S.G., Z.H., S.U. and I.A. helped conceive the basic idea, gave suggestions, G.L., Z.H. and S.G. arranged the outline of the content in the manuscript and prepared a draft article; corrected the entire article, and improved on the prospects in breeding programs. All authors read and approved the research article.

**Acknowledgments:** This work was financially supported by grants from the National Natural Science Foundation of China (31771350, 31471458) and the Key Science and Technology Support Program of Jiangsu Province (BE2018336) the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

**Conflicts of Interest:** The authors declare no conflict of interest.
