Detection of deleterious variants by protein language models is improved by nucleotide conservation at three evolutionary timescales in rice

This record contains the datasets, trained models, codes, and supporting materials for the manuscript “Detection of deleterious variants by protein language models is improved by nucleotide conservation at three evolutionary timescales in rice”. It includes standardized 'transformed' and 'non-transformed' training/validation/test datasets, prediction outputs for rice chromosome 12 (MAF validation), trained models, analysis scripts, and the conda environment file. See README.md for full details