Detection of QTNs for kernel moisture concentration and kernel dehydration rate before physiological maturity in maize using multi-locus GWAS

Abstract

Maize is China's largest grain crop. Mechanical grain harvesting is the key technology in maize production, and the kernel moisture concentration (KMC) is the main controlling factor in mechanical maize harvesting in China. The kernel dehydration rate (KDR) is closely related to the KMC. Thus, it is important to conduct genome-wide association studies (GWAS) of the KMC and KDR in maize, detect relevant quantitative trait nucleotides (QTNs), and mine relevant candidate genes. Here, 132 maize inbred lines were used to measure the KMC every 5 days from 10 to 40 days after pollination (DAP) in order to calculate the KDR. These lines were genotyped using a maize 55K single-nucleotide polymorphism array. QTNs for the KMC and KDR were detected based on five methods (mrMLM, FASTmrMLM, FASTmrEMMA, pLARmEB, and ISIS EM-BLASSO) in the package mrMLM. A total of 334 significant QTNs were found for both the KMC and KDR, including 175 QTNs unique to the KMC and 178 QTNs unique to the KDR; 116 and 58 QTNs were detected among the 334 QTNs by two and more than two methods, respectively; and 9 and 5 QTNs among 58 QTNs were detected in 2 and 3 years, respectively. A significant enrichment in cellular component was revealed by Gene Ontology enrichment analysis of candidate genes in the intervals adjacent to the 14 QTNs and this category contained five genes. The information provided in this study may be useful for further mining of genes associated with the KMC and KDR in maize.

Figure 1

The changes of the KMCs and KDRs in 2014, 2015 and 2016 across time.

Figure 2

Linkage disequilibrium decay and population structure of 132 lines. (a) LD decay; (b) Plot of delta K against putative K ranging from 2 to 10; (c) Stacked bar plot.

Figure 3

Genome-wide distribution of SNPs throughout the 132 lines’ genomes. The outermost box with the scale represents the 10 maize chromosomes. The red histogram represents the density of SNPs. The yellow, red and blue scatters represent significant QTNs for the KMC and KDR traits in 2014, 2015 and 2016, respectively.

Figure 4

Boxplots for validation of the 14 co-detected QTNs (a ~ n). For each QTN, the population was divided into two groups according to allele types. The X-axis represents the two alleles for each QTN, while the Y-axis corresponds to the phenotype.