Optimizing genomic selection for drought tolerance in common bean: Multi-trait models and environment-specific strategies
Year: 2026
Type: article
Abstract: Drought tolerance in common bean (Phaseolus vulgaris L.) is critical for food security in water-limited regions, yet breeding progress is hindered by complex trait inheritance and genotype-by-environment interactions. This study evaluated the efficacy of genomic selection (GS) models to identify optimal strategies for accelerating genetic gain. A diverse panel of 350 genotypes was phenotyped under well-watered and drought-stress conditions for grain yield, chlorophyll content, leaf relative water content (RWC), and above-ground biomass, and was genotyped using SNP markers. Genomic Best Linear Unbiased Prediction (GBLUP), Bayesian methods (BayesB and BayesC), and Random Forest models were compared using five-fold cross-validation. GBLUP delivered the most stable prediction accuracy across traits, whereas the Bayesian models excelled for traits governed by large-effect QTLs. Drought stress reduced prediction accuracy relative to well-watered conditions, reflecting greater environmental noise. However, incorporating secondary physiological traits (chlorophyll content and RWC) into multi-trait models significantly enhanced grain yield prediction accuracy under drought conditions. Although larger training populations and higher marker densities improved prediction accuracy, both exhibited diminishing returns beyond optimal thresholds. The low reliability of cross-environment predictions underscores the necessity of environment-specific training populations. These findings provide actionable strategies for deploying multi-trait GS models to accelerate the development of climate-resilient common bean cultivars. Key words: Common bean, drought tolerance, genomic prediction, Genomic Best Linear Unbiased Prediction (GBLUP), multi-trait models.
Language: English
Subfield: Genetics
Domain: Life Sciences
SDG: Zero hunger
Open Access status: diamond
OpenAlex ID: W7202265028