Predicting phenotypes from high-dimensional genomes using gradient boosting decision trees

xsxsGenomic selection (GS) is an emerging technique for predicting unknown phenotypes using genome-wide marker coverage, allowing the use of efficient computational models to select individuals with high phenotypic values as candidate breeding populations. However, GS remains challenging inefficient crop due limited size training populations, nature genotype-environment interactions, and complex interaction patterns between molecular markers. In this study, we ensemble learning algorithms construct gradient boosted decision tree (GBDT) achieve prediction from genotypic We trained GBDT wheat dataset compared predictive performance six other widely used models. The mean normalized discounted cumulative gain (MNDCG) method was evaluate ability each model values. results study show that: (1) Bayesian converge reach a steady-state only when sufficient number iterations are set. As increases, accuracy but efficiency decreases significantly. When 200,000 performed, five similar converges smooth state, their 7.60% better than overall, 70 times that model. (2) Overall, overall RRBLUP best, some traits, still had higher (3) influenced by subset markers, markers models, so reasonable genetic data appropriate could improve