Background: Repetitive DNA sequences accounts for over 80% of maize genome. Although simple sequence repeats (SSRs) account for only 0.03% of the genome, they have been widely used in maize genetic research and breeding as highly informative codominant DNA markers. The genome-wide distribution and polymorphism of SSRs are not well studied due to the lack of high-quality genome DNA sequence data. Results: In this study, using data from high-quality de novo-sequenced maize genomes of five representative maize inbred lines, we revealed that SSRs were more densely present in telomeric region than centromeric region, and were more abundant in genic sequences than intergenic sequences. On genic sequences, tri- and hexanucleotide motifs were more abundant in CDS sequence and some mono- and dinucleotide motifs were more abundant in UTR sequences. Median length and chromosomal density of SSRs were both narrowly range-bound, with median length of 14-18 bp and genome-wide average density of 3355.77 bp/Mbp. LTR-RTs of < 0.4 Mya had higher SSR density (4498-4992 bp/Mbp). The genome-specific and motif-specific SSR polymorphism were studied. Their potential breeding applications were discussed. Conclusions: We found that the median length of SSR sequences of different SSR motifs was nearly constant. SSR density in genic regions was much higher than intergenic regions. In addition, SSR density at LTR-RTs of different evolutionary ages varied in a narrow range. The SSRs and their LTR-RT carriers evolved at an equal rate. All these observations indicated that SSR length and density were under control of yet unknown evolutionary forces. The chromosome region-specific and motif-specific SSR polymorphisms we observed supported the notion that SSR polymorphism was invaluable genome resource for developing highly informative genome and gene markers in maize genetic research and molecular breeding.

Maize (Zea mays L.) germplasm in China Summer maize ecological region (CSM) or central corn-belt of China is diverse but has not been systematically characterized at molecular level. In this study, genetic variation, genome diversity, linkage disequilibrium patterns, population structure, and characteristics of different heterotic groups were studied using 525,141 SNPs obtained by Genotyping-By-Sequencing (GBS) for 490 inbred lines collected from researchers at CSM region. The SNP density is lower near centromere, but higher near telomere region of maize chromosome, the degree of linkage disequilibrium (r²) vary at different chromosome regions. Majority of the inbred lines (66.05%) show pairwise relative kinship near zero, indicating a large genetic diversity in the CSM breeding germplasm. Using 4849 tagSNPs derived from 3618 haplotype blocks, the 490 inbred lines were delineated into 3 supergroups, 6 groups, and 10 subgroups using ADMIXTURE software. A procedure of assigning inbred lines into heterotic groups using genomic data and tag-SNPs was developed and validated. Genome differentiation among different subgroups measured by F_st, and the genetic diversity within each subgroup measured by GD are both large. The share of heterotic groups that have significant North American germplasm contribution: P, SS, IDT, and X, accounts about 54% of the CSM breeding germplasm collection and has increased significantly in the last two decades. Two predominant types of heterotic pattern in CSM region are: M-Reid group × TSPT group, and X subgroup × Local subgroups.

Plant height (PH) and ear height (EH) are important traits associated with biomass, lodging resistance, and grain yield in maize. There were strong effects of genotype x environment interaction (GEI) on plant height and ear height of maize. In this study, 203 maize inbred lines were grown at five locations across China’s Spring and Summer corn belts, and plant height (PH) and ear height (EH) phenotype data were collected and grouped using GGE biplot. Five locations fell into two distinct groups (or mega environments) that coincide with two corn ecological zones called Summer Corn Belt and Spring Corn Belt. In total, 73,174 SNPs collected using GBS sequencing platform were used as genotype data and a recently released multi-environment GWAS software package IIIVmrMLM was employed to identify QTNs and QTN x environment (corn belt) interaction (QEIs); 12 and 11 statistically significant QEIs for PH and EH were detected respectively and their phenotypic effects were further partitioned into Add*E and Dom*E components. There were 28 and 25 corn-belt-specific QTNs for PH and EH identified, respectively. The result shows that there are a large number of genetic loci underlying the PH and EH GEIs and IIIVmrMLM is a powerful tool in discovering QTNs that have significant QTN-by-Environment interaction. PH and EH candidate genes were annotated based on transcriptomic analysis and haplotype analysis. EH related-QEI S10_135 (Zm00001d025947, saur76, small auxin up RNA76) and PH related-QEI S4_4 (Zm00001d049692, mads32, encoding MADS-transcription factor 32), and corn-belt specific QTNs including S10_4 (Zm00001d023333, sdg127, set domain gene127) and S7_1 (Zm00001d018614, GLR3.4, and glutamate receptor 3.4 or Zm00001d018616, DDRGK domain-containing protein) were reported, and the relationship among GEIs, QEIs and phenotypic plasticity and their biological and breeding implications were discussed.

Genetic dissection and breeding by design for polygenic traits remain challenges. To meet these challenges, it is important to identify as many genes as possible and key genes. Therefore, here, a genome-wide scanning plus machine learning framework was developed and integrated with advanced computational techniques to propose a novel algorithm called Fast3VmrMLM to mine more and key genes for polygenic traits in the era of big data and artificial intelligence. The algorithm was also extended to identify haplotype (Fast3VmrMLM-Hap) and molecular (Fast3VmrMLM-mQTL) variants. In simulation studies, Fast3VmrMLM outperformed existing methods in detecting dominant, small and rare variants, taking 3.30 and 5.43 hours (20 threads) to analyze the 18K rice and UK biobank-scale datasets, respectively. Fast3VmrMLM identified more known (211) and candidate (384) genes for 14 traits in the 18K rice dataset than FarmCPU (100 known genes), while Fast3VmrMLM identified 26 known and 24 candidate genes for 7 yield-related traits in a maize NC II design and Fast3VmrMLM-mQTL identified two known soybean genes around structural variants. We demonstrated that the new two-step framework outperformed genome-wide scanning alone. In breeding by design, a genetic network constructed by machine learning using all known/candidate genes in this study identified 21 key genes for rice yield-related traits, while all the associated markers gave high prediction accuracies in rice (0.7443) and maize (0.8492) and excellent hybrid combinations. A new breeding by design strategy based on the identified key genes was also proposed. This study provides an excellent method for gene mining and breeding by design.

Southern corn rust (SCR) caused by Puccinia polysora Underw is a major disease leading to severe yield losses in China Summer Corn Belt. Using six multi-locus GWAS methods, we identified a set of SCR resistance QTNs from a diversity panel of 140 inbred lines collected from China Summer Corn Belt. Thirteen QTNs on chromosomes 1, 2, 4, 5, 6, and 8 were grouped into three types of allele effects and their associations with SCR phenotypes were verified by post-GWAS case-control sampling, allele/haplotype effect analysis. Relative resistance (RR_R) and relative susceptibility (RRs) catering to its inbred carrier were estimated from single QTN and QTN-QTN combos and epistatitic effects were estimated for QTN-QTN combos. By transcriptomic annotation, a set of candidate genes were predicted to be involved in transcriptional regulation (S5_145, Zm00001d01613, transcription factor GTE4), phosphorylation (S8_123, Zm00001d010672, Pgk2- phosphoglycerate kinase 2), and temperature stress response (S6_164a/S6_164b, Zm00001d038806, hsp101, and S5_211, Zm00001d017978, cellulase25). The breeding implications of the above findings were discussed.

Although phenotypic plasticity facilitates the understanding of trait variation, its study remains limited. To study phenotypic plasticity, in this study, 203 inbred maize lines were phenotyped for flowering time and plant height in Zhengzhou, Ningjin, Tieling, and Binxian and genotyped with 121,871 high-quality SNPs. The regression slopes and intercepts of flowering time and plant height on four meteorological factors in their corresponding, most significant correlation windows were used to estimate the phenotypic plasticity of the above traits and to further indirectly detect the interactions between quantitative trait nucleotides and meteorological factors. Of the two known and seven candidate genes identified in this study, ZmCCT, GRMZM2G035417, GRMZM2G069651, and GRMZM2G359322 can be used to explain why spring maize has a longer flowering time than summer maize, as these genes delay pollen development and flowering regulators under long day and low temperature; while ZmPIN1b, GRMZM2G062045, GRMZM2G370777, GRMZM2G077752, and GRMZM2G126397 can be used to explain why Tieling has higher plant height than other regions, as these genes enhance auxin transport and suppress dwarfing genes under increased precipitation and low temperature. This study explains the phenotypic plasticity of these traits.