Effects of using deep learning to predict the geographic origin of barley genebank accessions on genome-environment association studies
Authors: Chang, C.-W., Schmid, K. J.
Date: 2025-02-17 · Version: 1
DOI: 10.1101/2025.02.13.638161 Category: Plant Biology
Model Organism: Hordeum vulgare
▶ AI Summary
The study applied neural networks to predict missing geographic origins of barley (Hordeum vulgare) accessions and incorporated the imputed environmental data into genome‑environment association (GEA) analyses. While the neural models achieved high cross‑validation accuracy, some predictions were ecologically implausible, and expanding sample size via imputation did not markedly improve GEA sensitivity, though it revealed complementary adaptive signals near flowering‑time genes.
Genome‑environment association Neural network geographic prediction Barley (Hordeum vulgare) Imputed environmental data Adaptive loci