Skip to main content

Table 1 Performance comparison of the different feature selection strategies

From: m5U-SVM: identification of RNA 5-methyluridine modification sites based on multi-view features of physicochemical features and distributed representation

Mode

Feature selection

Acc (%)

AUC

Sn (%)

Sp (%)

MCC

F1

Full transcript

ANOVA-IFS

87.853

0.9422

81.356

91.335

0.7312

0.8237

F-score-IFS

87.648

0.9421

81.095

91.160

0.7267

0.8209

LightGBM -IFS

88.740

0.9504

80.574

93.117

0.7493

0.8332

XGBoost-IFS

88.376

0.9461

80.117

92.802

0.7412

0.8279

Mature mRNA

ANOVA-IFS

93.699

0.9756

92.879

94.518

0.8741

0.9364

F-score-IFS

93.699

0.9756

92.879

94.518

0.8741

0.9363

LightGBM -IFS

94.309

0.9792

93.286

95.330

0.8864

0.9425

XGBoost-IFS

93.902

0.9735

92.777

95.025

0.8783

0.9383