Table 11
Open in new tab

Component ablation: mean MAE of nine torsion angles

(a) RNA-TorsionBERT Dataset
Model VariantMean MAE (⁠|$^\circ $|⁠)
Full Model19.78
Without Transformer Encoder22.90
Without 2A3IDC Modules24.31
Without Multi-head Attention21.86
(b) SPOT-RNA-1D Dataset
Model VariantMean MAE (⁠|$^\circ $|⁠)
Full Model23.58
Without Transformer Encoder26.34
Without 2A3IDC Modules28.72
Without Multi-head Attention25.91
(a) RNA-TorsionBERT Dataset
Model VariantMean MAE (⁠|$^\circ $|⁠)
Full Model19.78
Without Transformer Encoder22.90
Without 2A3IDC Modules24.31
Without Multi-head Attention21.86
(b) SPOT-RNA-1D Dataset
Model VariantMean MAE (⁠|$^\circ $|⁠)
Full Model23.58
Without Transformer Encoder26.34
Without 2A3IDC Modules28.72
Without Multi-head Attention25.91
Table 11
Open in new tab

Component ablation: mean MAE of nine torsion angles

(a) RNA-TorsionBERT Dataset
Model VariantMean MAE (⁠|$^\circ $|⁠)
Full Model19.78
Without Transformer Encoder22.90
Without 2A3IDC Modules24.31
Without Multi-head Attention21.86
(b) SPOT-RNA-1D Dataset
Model VariantMean MAE (⁠|$^\circ $|⁠)
Full Model23.58
Without Transformer Encoder26.34
Without 2A3IDC Modules28.72
Without Multi-head Attention25.91
(a) RNA-TorsionBERT Dataset
Model VariantMean MAE (⁠|$^\circ $|⁠)
Full Model19.78
Without Transformer Encoder22.90
Without 2A3IDC Modules24.31
Without Multi-head Attention21.86
(b) SPOT-RNA-1D Dataset
Model VariantMean MAE (⁠|$^\circ $|⁠)
Full Model23.58
Without Transformer Encoder26.34
Without 2A3IDC Modules28.72
Without Multi-head Attention25.91
Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close