Data and Results: Protein Language Model Embeddings for HIV Drug Resistance Prediction

Raw genotype-phenotype data, processed datasets, analysis results, and publication-quality figures for the study "Protein Language Model Embeddings Improve HIV Drug Resistance Prediction: A Comprehensive Benchmark with Attention-Based Interpretability." Includes Stanford HIVDB genotype-phenotype datasets (PI, NRTI, NNRTI), subtype assignments, multi-PLM comparison results (ESM-2, ESM C 600M, ESM-1v), subtype-stratified performance, temporal holdout validation, and all 10 manuscript figures at 400 DPI. Analysis code is archived separately on Zenodo (DOI: 10.5281/zenodo.19466630).