Skip to main content

Table 11 The performance of different modules on HLA-G*01:04 dataset

From: DeepHLAPred: a deep learning-based method for non-classical HLA binder prediction

HLA-G*01:04

SN

SP

ACC

MCC

AUC

Model

     

 CNN

0.9264

0.9365

0.933

0.8627

0.9782

 Bi-LSTM

0.9236

0.9234

0.9236

0.8475

0.9729

 CNN + Bi-LSTM (In series)

0.9457

0.9394

0.9386

0.8858

0.9836

 CNN + CNN (In parallel)

0.9310

0.9396

0.9354

0.8727

0.9802

 Bi-LSTM + Bi-LSTM (In parallel)

0.9211

0.9434

0.9323

0.8657

0.9760

 DeepHLAPred

0.9545

0.9630

0.9587

0.9179

0.9855