Skip to main content

Table 6 Comparison of the Se, Sp, ACC and MCC between four traditional models and sequence information model on Dataset 2

From: Accurate classification of membrane protein types based on sequence and evolutionary information using deep learning

Membrane protein types

Index

AAC

DipC

TipC

PseAAC

Our model

 

Se

0.6571

0.5306

0.3388

0.6898

0.8408

Type I

Sp

0.9384

0.9643

0.9824

0.9440

0.9806

 

ACC

0.9193

0.9348

0.9387

0.9267

0.9711

 

MCC

0.4951

0.4903

0.4156

0.5341

0.7841

 

Se

0.1139

0.0

0.0

0.0633

0.7468

Type II

Sp

0.9901

1.0

1.0

0.9889

0.9739

 

ACC

0.9709

0.9781

0.9781

0.9686

0.9689

 

MCC

0.1387

0.0

0.0

0.0696

0.5267

 

Se

0.0

0.0

0.0

0.0

0.0

Type III

Sp

1.0

1.0

1.0

1.0

1.0

 

ACC

0.9975

0.9975

0.9975

0.9975

0.9975

 

MCC

0.0

0.0

0.0

0.0

0.0

 

Se

0.0

0.0

0.0

0.0

0.7647

Type IV

Sp

1.0

1.0

1.0

1.0

0.9958

 

ACC

0.9953

0.9953

0.9953

0.9953

0.9947

 

MCC

0.0

0.0

0.0

0.0

0.5935

 

Se

0.8721

0.9290

0.9617

0.8709

0.9157

Mutipass

Sp

0.8526

0.7105

0.4885

0.8588

0.9813

 

ACC

0.8660

0.8607

0.8138

0.8671

0.9362

 

MCC

0.7022

0.6669

0.5442

0.7058

0.8638

 

Se

0.0556

0.0

0.0

0.0278

0.0

Lipid-chain-anchored

Sp

1.0

1.0

1.0

1.0

0.9983

 

ACC

0.9906

0.9900

0.9900

0.9903

0.9883

 

MCC

0.2346

0.0

0.0

0.1659

-0.0041

 

Se

0.3659

0.0488

0.0

0.3659

0.9024

GPI-anchored

Sp

0.9986

1.0

1.0

0.9941

0.9938

 

ACC

0.9914

0.9892

0.9886

0.9870

0.9928

 

MCC

0.5203

0.2197

0.0

0.3839

0.7490

 

Se

0.7425

0.6881

0.5451

0.7797

0.9013

Peripheral

Sp

0.8885

0.9164

0.9580

0.8954

0.9418

 

ACC

0.8602

0.8721

0.8779

0.8729

0.9340

 

MCC

0.5893

0.5965

0.5740

0.6290

0.8025

  1. Note: Best performing method in bold.