Genomic language models improve cross-species gene expression prediction and accurately capture regulatory variant effects in Brachypodium mutant lines

This record contains the datasets, trained models, codes, and supporting materials for the manuscript “Genomic language models improve cross-species gene expression prediction and accurately capture regulatory variant effects in Brachypodium mutant lines”. It contains datasets including sequence embeddings across 17 plant species used for training, hyperparameter optimization, cross-validation tests, and experimental in planta validation of predictions, gene expression predictions across 17 species by each of the 4 EMPRES model types, trained models, analysis scripts, and the conda environment file. See README.md for full details.