Dissecting the genetic basis of agronomic traits by multi-trait GWAS and genetic networks in maize (Zea mays L.)

This study is led by Professors Huihui Li (Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, Beijing, China), Liang Li (Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, Beijing, China), and Enying Zhang (Qingdao Agricultural University, Qingdao, China). The authors systematically analyzed phenotypic variation across three ecological environments using 2,448 maize inbred lines. By integrating single-trait GWAS, multi-trait joint analysis (MTAG), and genetic network construction, we decoded the genetic architecture of 18 agronomic traits, identified pleiotropic regulatory genes, and revealed interconnected co-regulatory networks.

Single-trait GWAS using the FarmCPU model identified 558 significant SNPs associated with 18 traits, corresponding to 457 candidate genes. Subsequent MTAG revealed 546 significant SNPs, including 182 novel loci not detected in the single-trait analysis. These novel candidate genes exhibited cross-trait associations and were enriched in key biological processes such as cell division, hormone signaling, and metabolic synthesis, suggesting their coordinated regulatory roles in plant architecture, flowering time, and yield formation.

A genetic interaction network integrating GWAS/MTAG results was constructed, where nodes represent QTLs/traits and edges denote linkage disequilibrium. Key pleiotropic regions included the QTL411 locus containing kn1 (regulating plant/ear height), which interacted with ZmGA20ox1 and rth3. We identified 49 "hub QTLs" connecting ≥2 traits, providing strategic targets for simultaneous trait improvement.

RT-qPCR analysis of 30 candidate gene pairs across 12 tissues at two developmental stages (V12: flare opening stage; VT: tasseling stage) confirmed synchronized co-expression: Plant/ear height genes showed co-upregulation during vegetative growth (V12), while flowering-time genes exhibited coordinated activation at reproductive transition (VT). These spatiotemporal patterns aligned with genetic interactions, revealing dynamic co-regulation mechanisms.

This work establishes an integrated GWAS-MTAG-network framework, deciphers the genetic interaction landscape of maize agronomic traits, and provides theoretical foundations for leveraging pleiotropic genes in molecular design breeding.

See the article:

Dissecting the genetic basis of agronomic traits by multi-trait GWAS and genetic networks in maize (Zea mays L.)

https://link.springer.com/article/10.1007/s42994-025-00241-4