Protein Property Prediction on Protein Property Datasets (Unseen mutations)
0.629Spearman CorrelationKermut-GP
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Kermut-GPModel Class=Gaussian process, Prior Information=ESM-2 + ProteinMPNN, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.629 | 0.632 | 0.617 | |
| KermutStruc-GPModel Class=Gaussian process, Prior Information=ESM-2 + ProteinMPNN, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.614 | 0.613 | 0.634 | |
| LOCK-GPModel Class=Gaussian process, Prior Information=BLOSUM, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.61 | 0.622 | 0.591 | |
| MLP-ESM2-LastLayerModel Class=Neural network, Prior Information=ESM-2, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.558 | 0.573 | 0.626 | |
| Tanimoto-GPModel Class=Gaussian process, Prior Information=BLOSUM, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.555 | 0.56 | 0.675 | |
| KermutSeq-GPModel Class=Gaussian process, Prior Information=ESM-2, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.532 | 0.537 | 0.681 | |
| SigGLM-OHModel Class=Linear + global non-linearity, Prior Information=—, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.515 | 0.535 | 0.661 | |
| Ridge-OHModel Class=Linear, Prior Information=—, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.514 | 0.519 | 0.682 | |
| Ridge-ESM2Model Class=Neural network, Prior Information=ESM-2, Number of training data points=96, Evaluation regime=Unseen mutations2026.06 | 0.476 | 0.49 | 1.628 |