Skip to main content

Table 5 Performance of different models in predicting miRNA-abiotic stress associations under five-fold cross-validation

From: Predicting abiotic stress-responsive miRNA in plants based on multi-source features fusion and graph neural network

Model

AUPR

AUC

F1

ACC

RE

SPE

PRE

GIN

0.9824

0.9743

0.9495

0.9499

0.9453

0.9544

0.9545

GraphSAGE

0.9726

0.9612

0.9348

0.9347

0.9368

0.9325

0.9334

GCN

0.9494

0.9306

0.8882

0.8913

0.8639

0.9186

0.9143

GAT

0.9136

0.9127

0.8843

0.8837

0.8870

0.8803

0.8822

SVM

0.9615

0.9733

0.9328

0.9320

0.9526

0.9117

0.9142

KNN

0.9447

0.9464

0.8652

0.8548

0.9364

0.7751

0.8052

RF

0.9442

0.9383

0.8478

0.8478

0.8568

0.8376

0.8400

  1. F1, F1 score; ACC, accuracy; RE, recall; SPE, specificity;PRE, precision, Bold indicates the maximum value of the column
Back to article page

Plant Methods

ISSN: 1746-4811