BERT-PPII: The Polyproline Type II Helix Structure Prediction Model Based on BERT and Multichannel CNN

<div>(a) The ROC plot of the BRET-PPII model on the Strict independent test set; (b) The ROC plot of the BERT-PPII model on the NonStrict independent test set. TPR represents the rate that is correctly judged to be positive, and FPR represents the rate that is wrongly judged to be positive.</div>

BioMed Research International

fig8

Figure 8

Figure 8: BERT-PPII: The Polyproline Type II Helix Structure Prediction Model Based on BERT and Multichannel CNN 