By analyzing 1771 RNA-seq datasets from seven tissues in a maize diversity panel, we explored the landscape of multi-tissue transcriptome variation and evolution patterns of tissue-specific genes, and built a comprehensive multi-tissue gene regulation atlas to understand the genetic regulation of maize complex trait. Using transcriptome-wide association analysis, we linked tissue-specific expression variation of 45 genes to variation of 11 agronomic traits. Through integrative analyses of tissue-specific gene regulatory variation with genome-wide association studies, we detected relevant tissue types and candidate genes for a number of agronomic traits, including leaf during the day for anthesis-silking interval (GRMZM2G093210), leaf during the day for kernel Zeinoxanthin level (GRMZM2G143202), and root for ear height (GRMZM2G700665), highlighting the contribution from tissue-specific gene expression to variation of agronomic trait. Our findings provide novel insights into the genetic and biological mechanisms underlying complex traits in maize, and the multi-tissue regulatory atlas serves as a primary source for biological interpretation, functional validation, and genomic improvement of maize.

Zinc (Zn) malnutrition is a major public health issue. Genetic biofortification of Zn in rice grain can alleviate global Zn malnutrition. Therefore, elucidating the genetic mechanisms regulating Zn deprivation response in rice is essential to identify elite genes useful for breeding high grain Zn rice varieties. Here, a meta-analysis of previous RNA-Seq studies involving Zn deficient conditions was conducted using the weighted gene co-expression network analysis (WGCNA) and other in silico prediction tools to identify modules (denoting cluster of genes with related expression pattern) of co-expressed genes, modular genes which are conserved differentially expressed genes (DEGs) across independent RNA-Seq studies, and the molecular pathways of the conserved modular DEGs. WGCNA identified 16 modules of co-expressed genes. Twenty-eight and five modular DEGs were conserved in leaf and crown, and root tissues across two independent RNA-Seq studies. Functional enrichment analysis showed that 24 of the 28 conserved modular DEGs from leaf and crown tissues significantly up-regulated 2 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways and 15 Gene Ontology (GO) terms, including the substrate- specific transmembrane transporter and the small molecule metabolic process. Further, the well-studied transcription factors (OsWOX11 and OsbHLH120), protein kinase (OsCDPK20 and OsMPK17), and miRNAs (OSA-MIR397A and OSA-MIR397B) were predicted to target some of the identified conserved modular DEGs. Out of the 24 conserved and up-regulated modular DEGs, 19 were yet to be experimentally validated as Zn deficiency responsive genes. Findings from this study provide a comprehensive insight on the molecular mechanisms of Zn deficiency response and may facilitate gene and pathway prioritization for improving Zn use efficiency and Zn biofortification in rice.

Multi-trait (MT) genomic prediction models enable breeders to save phenotyping resources and increase the prediction accuracy of unobserved target traits by exploiting available information from non-target or auxiliary traits. Our study evaluated different MT models using 250 rice accessions from Asian countries genotyped and phenotyped for grain content of zinc (Zn), iron (Fe), copper (Cu), manganese (Mn), and cadmium (Cd). The predictive performance of MT models compared to a traditional single trait (ST) model was assessed by 1) applying different cross-validation strategies (CV1, CV2, and CV3) inferring varied phenotyping patterns and budgets; 2) accounting for local epistatic effects along with the main additive effect in MT models; and 3) using a selective marker panel composed of trait-associated SNPs in MT models. MT models were not statistically significantly (p < 0.05) superior to ST model under CV1, where no phenotypic information was available for the accessions in the test set. After including phenotypes from auxiliary traits in both training and test sets (MT-CV2) or simply in the test set (MT-CV3), MT models significantly (p < 0.05) outperformed ST model for all the traits. The highest increases in the predictive ability of MT models relative to ST models were 11.1% (Mn), 11.5 (Cd), 33.3% (Fe), 95.2% (Cu) and 126% (Zn). Accounting for the local epistatic effects using a haplotype-based model further improved the predictive ability of MT models by 4.6% (Cu), 3.8% (Zn), and 3.5% (Cd) relative to MT models with only additive effects. The predictive ability of the haplotype-based model was not improved after optimizing the marker panel by only considering the markers associated with the traits. This study first assessed the local epistatic effects and marker optimization strategies in the MT genomic prediction framework and then illustrated the power of the MT model in predicting trace element traits in rice for the effective use of genetic resources to improve the nutritional quality of rice grain.

Mesocotyl is a crucial organ for pushing buds out of soil, which plays a vital role in seedling emergence and establishment in direct-seeded rice. Thus, the identification of quantitative trait loci (QTL) associated with mesocotyl length (ML) could accelerate genetic improvement of rice for direct seeding cultivation. In this study, QTL sequencing (QTL-seq) applied to 12 F2 populations identified 14 QTL for ML, which were distributed on chromosomes 1, 3, 4, 5, 6, 7, and 9 based on the Δ(SNP-index) or G-value statistics. Besides, a genome-wide association study (GWAS) using two diverse panels identified five unique QTL on chromosomes 1, 8, 9, and 12 (2), respectively, explaining 5.3–14.6% of the phenotypic variations. Among these QTL, seven were in the regions harboring known genes or QTLs, whereas the other 10 were potentially novel. Six of the QTL were stable across two or more populations. Eight high-confidence candidate genes related to ML were identified for the stable loci based on annotation and expression analyses. Association analysis revealed that two PCR gel-based markers for the loci co-located by QTL-seq and GWAS, Indel-Chr1:18932318 and Indel-Chr7:15404166 for loci qML1.3 and qML7.2 respectively, were significantly associated with ML in a collection of 140 accessions and could be used as breeder-friendly markers in further breeding.

Salinity is one of the most important abiotic stresses, which seriously affects rice production. In this study, a multiparent advanced generation intercross population, DC1, derived from intercrossing four elite indica varieties, was used to identify QTLs conferring salt tolerance. The whole population and four parents were genotyped with a 55K rice SNP array. A total of 7 QTLs delineated from 186 significant marker-trait associations were detected on chromosomes 1, 2, 5 and 9, which accounted for 7.42–9.38% of the total phenotypic variations. Among these QTLs, one novel QTL (qRRL2) on chromosome 2 for relative root length and one multi-trait QTL (qSLST1/qRDSW1/qRB1) on chromosome 1 affecting shoot length, root dry weight and root biomass under salt treatment were detected. Gene expression analysis revealed that a transcription factor gene (LOC_Os01g66280) within the multi-trait QTL is the potential candidate gene for salt tolerance. Interestingly, we identified one significant nonsynonymous SNP in the coding region of this candidate gene. These results will facilitate fine mapping of the candidate gene and QTL pyramiding to genetically improve salt tolerance in rice.