Integrated GWAS and linkage mapping in a multi-parent population prioritizes candidate genes for kernel number per row in maize

Background Kernel number per row (KNPR) is a key component of maize yield. The molecular mechanisms and favorable allelic variants underlying KNPR variation remain poorly characterized in temperate non-Reid germplasm. This study aimed to improve mapping resolution and prioritize candidate loci associated with KNPR using a multi-parent population. Results We established a multi-parent population comprising 530 recombinant inbred lines derived from crosses between the Reid inbred Ye107 and four genetically distinct elite lines. An integrated analytical pipeline combining genome-wide association analysis, linkage QTL mapping, haplotype analysis, protein-structure modeling, and spatiotemporal expression profiling was used to identify KNPR-associated regions and prioritize candidate genes. An exploratory GEBV analysis was conducted separately to describe model-derived genomic-value patterns. Four candidate genes were prioritized as testable hypotheses: ZmNMD3 (ribosomal nuclear export), ZmPAPS1 (mRNA processing), ZmNAC70 (NAC transcription factor), and ZmF-box1 (F-box protein). Conclusions Multi-parent populations can facilitate the discovery of allelic variation in temperate non-Reid maize germplasm. The candidate genes and haplotypes prioritized here provide testable hypotheses for future functional validation. Favorable effects confirmed in replicated, multi-environment experiments would justify subsequent evaluation of targeted introgression and multi-gene pyramiding for precision maize breeding.