TargetTrackHisTag: benchmark dataset and ESM-2 embeddings for His-tag placement prediction

ESM-2 embeddings, labels, benchmark dataset, and analysis results for TargetTrackHisTag — the first machine-readable experimental benchmark of bilateral His-tag placement outcomes curated from the NESG TargetTrack archive (156 proteins). Files:- experimental_156_clean.csv: Core benchmark (156 proteins, experimental labels)- target_ids_156.csv: TargetTrack identifiers- X_exp_156_esm.npy: ESM-2 (650M) full-sequence embeddings (156 × 1280)- y_exp_156.npy: Binary labels (1=N-winner, 0=C-winner)- results/: LOOCV predictions, permutation test, length analysis outputs- audit_results/: Dataset quality audit outputs (Audits A, B, C)