Skip to main content

Table 2 Evaluation based on four singer filters and LRMR-Ri for membrane protein dataset with three models

From: Predicting antifreeze proteins with weighted generalized dipeptide composition and multi-regression feature selection ensemble

Model

Evaluating indicator

Lasso

Ridge

Mic

Relief

LRMR-Ri

DT

PE

0.3296

0.4928

0.3487

0.6778

0.4585

RE

0.3350

0.3760

0.3425

0.4376

0.4258

F-Measure

0.3316

0.4021

0.3443

0.4339

0.4353

ACC (%)

69.86

77.97

71.82

77.59

79.83

MCC

0.5037

0.6101

0.5036

0.6171

0.6563

SVM

PE

0.5446

0.5352

0.5228

0.5569

0.5159

RE

0.4853

0.4391

0.4762

0.4574

0.4432

F-Measure

0.5051

0.4616

0.4910

0.4842

0.4549

ACC (%)

65.68

67.41

67.70

68.43

72.57

MCC

0.5133

0.5766

0.5601

0.5951

0.6007

SGD

PE

0.3487

0.5777

0.6304

0.4776

0.5593

RE

0.3425

0.4614

0.3911

0.3847

0.4460

F-Measure

0.3443

0.4934

0.4384

0.4076

0.4810

ACC (%)

71.82

78.19

75.88

75.88

80.15

MCC

0.5237

0.6304

0.5648

0.5827

0.6562