Genome-wide association analysis and prediction model construction for soybean plant height

Plant height is a key agronomic trait that is closely associated with soybean yield potential and lodging resistance, and it is essential for achieving high and stable yields in soybean production. Here, we conducted a genome-wide association study (GWAS) and genomic selection (GS) analyses of plant height using a diverse panel of 17,148 soybean germplasm accessions with worldwide geographic representation. We identified 139 significant genomic loci associated with plant height, with the candidate gene Dt1 showing the largest genetic effect. In the GS analysis, the BayesA and rrBLUP models outperformed the other models in prediction accuracy. Further comparisons indicated that prediction models based on the top 500-1000 SNPs provided the best balance between predictive performance and cost effectiveness. Collectively, the loci and models identified here support efficient genomic selection for soybean plant height and provide a theoretical basis for molecular design breeding to improve this trait.