Protein Property Prediction on 21 Protein Property Datasets (Cross-validation)
0.867Spearman CorrelationLOCK-GP
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| LOCK-GPModel Class=Gaussian process, Prior Information=BLOSUM, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.867 | 0.914 | 0.21 | |
| MLP-ESM2-LastLayerModel Class=Neural network, Prior Information=ESM-2, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.851 | 0.9 | 0.243 | |
| Kermut-GPModel Class=Gaussian process, Prior Information=ESM-2 + ProteinMPNN, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.85 | 0.888 | 0.285 | |
| Tanimoto-GPModel Class=Gaussian process, Prior Information=BLOSUM, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.846 | 0.888 | 0.272 | |
| Ridge-ESM2Model Class=Neural network, Prior Information=ESM-2, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.837 | 0.877 | 0.301 | |
| KermutStruc-GPModel Class=Gaussian process, Prior Information=ESM-2 + ProteinMPNN, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.809 | 0.805 | 0.417 | |
| SigGLM-OHModel Class=Linear + global non-linearity, Prior Information=—, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.804 | 0.849 | 0.342 | |
| Ridge-OHModel Class=Linear, Prior Information=—, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.802 | 0.8 | 0.422 | |
| KermutSeq-GPModel Class=Gaussian process, Prior Information=ESM-2, Number of training data points=1536, Evaluation regime=Cross-validation2026.06 | 0.794 | 0.838 | 0.338 |