Table 5 Performance comparison of cold-start for disease. The best performance is denoted as bold.

From: Decoding potential lncRNA and disease associations through graph representation learning and gradient boosting with histogram

Metric

Dataset

SDLDA

LDNFSGB

IPCARF

LDA-VGHB

LDA-GMCB

Precision

lncRNADisease v2.0

0.8854 ± 0.0377

0.7548 ± 0.0639

0.5583 ± 0.0910

0.8917 ± 0.0316

0.8836 ± 0.0356

MNDR

0.9232 ± 0.0331

0.8005 ± 0.0625

0.5557 ± 0.1473

0.9300 ± 0.0251

0.9282 ± 0.0310

Recall

lncRNADisease v2.0

0.7182 ± 0.0694

0.7309 ± 0.0646

0.7538 ± 0.1067

0.8415 ± 0.0449

0.8386 ± 0.0450

MNDR

0.8579 ± 0.0655

0.6936 ± 0.0794

0.5279 ± 0.1969

0.9190 ± 0.0397

0.9033 ± 0.0472

Accuracy

lncRNADisease v2.0

0.8187 ± 0.0282

0.7552 ± 0.0291

0.5766 ± 0.0740

0.8737 ± 0.0177

0.8674 ± 0.0254

MNDR

0.9043 ± 0.0174

0.7670 ± 0.0432

0.5593 ± 0.1159

0.9305 ± 0.0153

0.9238 ± 0.0180

F1-score

lncRNADisease v2.0

0.7917 ± 0.0519

0.7407 ± 0.0526

0.6339 ± 0.0715

0.8651 ± 0.0304

0.8599 ± 0.0340

MNDR

0.8886 ± 0.0475

0.7402 ± 0.0577

0.5190 ± 0.1434

0.9242 ± 0.0298

0.9152 ± 0.0366

AUC

lncRNADisease v2.0

0.8788 ± 0.0274

0.8329 ± 0.0273

0.6402 ± 0.1004

0.9406 ± 0.0154

0.9411 ± 0.0177

MNDR

0.9559 ± 0.0160

0.8603 ± 0.0363

0.5992 ± 0.1601

0.9741 ± 0.0106

0.9751 ± 0.0125

AUPR

lncRNADisease v2.0

0.8934 ± 0.0387

0.8163 ± 0.0537

0.6355 ± 0.1217

0.9429 ± 0.0233

0.9474 ± 0.0202

MNDR

0.9561 ± 0.0354

0.8292 ± 0.0680

0.6040 ± 0.1476

0.9728 ± 0.0204

0.9731 ± 0.0267

Back to article page