Skip to main content

Table 5 Performance of several advanced models on independent test sets

From: EMDL_m6Am: identifying N6,2′-O-dimethyladenosine sites based on stacking ensemble deep learning

model

Sn

Sp

MCC

ACC

AUC

AUPR

Pre

F1

VGG-16

0.9662

0.1634

0.2173

0.5648

0.7643

0.7546

0.5359

0.6894

ResNet

0.6085

0.5859

0.1944

0.5972

0.6549

0.6500

0.5951

0.6017

CSPNet

0.7606

0.7155

0.4765

0.7380

0.8135

0.7094

0.7278

0.7438

VGG-19

0.5465

0.6986

0.2480

0.6225

0.6830

0.6604

0.6445

0.5915

Inception V3

0.5606

0.6423

0.2035

0.6014

0.6510

0.6482

0.6104

0.5844

EMDL_m6Am

0.8225

0.7972

0.6199

0.8098

0.8211

0.7626

0.8061

0.7960

  1. The best outcomes are in bold
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us