Skip to main content

Table 12 The performance of different modules on HLA-E*01:01 dataset

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

HLA-E*01:01

SN

SP

ACC

MCC

AUC

Model

     

 CNN

0.8497

0.8512

0.8521

0.7050

0.9540

 Bi-LSTM

0.8666

0.852

0.8588

0.7174

0.9209

 CNN + Bi-LSTM (In series)

0.8475

0.8730

0.8629

0.7234

0.9356

 CNN + CNN (In parallel)

0.9012

0.8588

0.8802

0.7605

0.9509

 Bi-LSTM + Bi-LSTM (In parallel)

0.8721

0.8665

0.8690

0.7378

0.9391

 DeepHLAPred

0.9413

0.9013

0.9226

0.8455

0.9595