Skip to main content

Table 7 The performance of five LPI prediction methods on CV3

From: LPI-deepGBDT: a multiple-layer deep framework based on gradient boosting decision trees for lncRNA–protein interaction identification

Metric

Dataset

LPI-BLS

LPI-CatBoost

PLIPCOM

LPI-SKF

LPI-HNM

LPI-deepGBDT

Precision

Dataset 1

0.8539 ± 0.0012

0.8340 ± 0.0170

0.8440 ± 0.0045

0.7979 ± 0.0337

0.7192 ± 0.0076

0.8572 ± 0.0143

Dataset 2

0.8668 ± 0.0018

0.8191 ± 0.0224

0.8478 ± 0.0021

0.7902 ± 0.0059

0.7104 ± 0.0081

0.8638 ± 0.0089

Dataset 3

0.7142 ± 0.0005

0.7349 ± 0.0183

0.7182 ± 0.0138

0.7631 ± 0.0095

0.7052 ± 0.0055

0.7565 ± 0.0313

Dataset 4

0.7012 ± 0.0065

0.6289 ± 0.0277

0.7498 ± 0.0144

0.7948 ± 0.0070

0.6527 ± 0.0124

0.8085 ± 0.0230

Dataset 5

0.7971 ± 0.0031

0.7425 ± 0.0047

0.7761 ± 0.0016

0.8248 ± 0.0011

0.8069 ± 0.0032

0.8578 ± 0.0066

Ave.

0.7866

0.7518

0.7872

0.7942

0.7189

0.8287

Recall

Dataset 1

0.6565 ± 0.0083

0.8308 ± 0.0154

0.9652 ± 0.0080

0.9379 ± 0.0283

0.6811 ± 0.0043

0.9684 ± 0.0071

Dataset 2

0.6603 ± 0.0068

0.8451 ± 0.0242

0.9504 ± 0.0012

0.6910 ± 0.0092

0.6485 ± 0.0116

0.9611 ± 0.0137

Dataset 3

0.6313 ± 0.0075

0.6951 ± 0.0336

0.7612 ± 0.0237

0.6745 ± 0.0065

0.6712 ± 0.0062

0.7588 ± 0.0939

Dataset 4

0.6445 ± 0.0046

0.5863 ± 0.0638

0.6988 ± 0.0143

0.7007 ± 0.0052

0.6177 ± 0.0162

0.7903 ± 0.0650

Dataset 5

0.7194 ± 0.0014

0.8691 ± 0.0035

0.8659 ± 0.0030

0.7304 ± 0.0006

0.6787 ± 0.0025

0.9003 ± 0.0151

Ave.

0.6624

0.7652

0.8483

0.7469

0.6594

0.8745

Accuracy

Dataset 1

0.7604 ± 0.0027

0.8319 ± 0.0170

0.8933 ± 0.0020

0.8488 ± 0.0136

0.6521 ± 0.0067

0.8877 ± 0.0075

Dataset 2

0.7687 ± 0.0032

0.8264 ± 0.0107

0.8976 ± 0.0018

0.6965 ± 0.0057

0.6439 ± 0.0087

0.9570 ± 0.0125

Dataset 3

0.6635 ± 0.0038

0.7194 ± 0.0061

0.7302 ± 0.0044

0.6745 ± 0.0065

0.6462 ± 0.0048

0.7683 ± 0.0136

Dataset 4

0.6542 ± 0.0044

0.6095 ± 0.0138

0.7322 ± 0.0092

0.7007 ± 0.0052

0.5958 ± 0.0107

0.8047 ± 0.0204

Dataset 5

0.7428 ± 0.0030

0.7837 ± 0.0030

0.8081 ± 0.0010

0.7304 ± 0.0006

0.7193 ± 0.0017

0.9355 ± 0.0028

Ave.

0.7179

0.7542

0.8123

0.7302

0.6515

0.8583

F1-score

Dataset 1

0.7421 ± 0.0048

0.8315 ± 0.0082

0.9005 ± 0.0020

0.8614 ± 0.0077

0.6996 ± 0.0055

0.8954 ± 0.0061

Dataset 2

0.7495 ± 0.0051

0.8295 ± 0.0094

0.9044 ± 0.0016

0.6565 ± 0.0071

0.6780 ± 0.0093

0.9200 ± 0.0101

Dataset 3

0.6702 ± 0.0019

0.7110 ± 0.0095

0.7379 ± 0.0043

0.6359 ± 0.0072

0.6878 ± 0.0045

0.8269 ± 0.0297

Dataset 4

0.6716 ± 0.0054

0.5881 ± 0.0264

0.7226 ± 0.0091

0.6636 ± 0.0057

0.6347 ± 0.0142

0.8042 ± 0.0306

Dataset 5

0.7563 ± 0.0022

0.8007 ± 0.0020

0.8186 ± 0.0011

0.6923 ± 0.0007

0.7373 ± 0.0015

0.8784 ± 0.0041

Ave.

0.7179

0.7521

0.8168

0.7019

0.6875

0.8429

AUC

Dataset 1

0.9247 ± 0.0012

0.8846 ± 0.0060

0.9292 ± 0.0016

0.9293 ± 0.0120

0.7800 ± 0.0108

0.9354 ± 0.0072

Dataset 2

0.9352 ± 0.0011

0.8918 ± 0.0055

0.9389 ± 0.0015

0.8893 ± 0.0136

0.7599 ± 0.0134

0.9423 ± 0.0060

Dataset 3

0.7883 ± 0.6735

0.7940 ± 0.0049

0.8229 ± 0.0025

0.8493 ± 0.0130

0.7693 ± 0.0083

0.8526 ± 0.0116

Dataset 4

0.7823 ± 0.0069

0.6421 ± 0.0122

0.8047 ± 0.0095

0.9024 ± 0.0105

0.6824 ± 0.0236

0.8542 ± 0.0137

Dataset 5

0.8826 ± 0.0031

0.8156 ± 0.0020

0.8903 ± 0.0010

0.9609 ± 0.0013

0.8874 ± 0.0029

0.9523 ± 0.0012

Ave.

0.8626

0.8056

0.8772

0.9062

0.7758

0.9073

AUPR

Dataset 1

0.8852 ± 0.0006

0.8904 ± 0.0084

0.9208 ± 0.0028

0.9290 ± 0.0155

0.8297 ± 0.0084

0.9043 ± 0.0162

Dataset 2

0.9013 ± 0.0035

0.8926 ± 0.0049

0.9049 ± 0.0028

0.8956 ± 0.0128

0.7897 ± 0.0120

0.9242 ± 0.0171

Dataset 3

0.7520 ± 0.0006

0.7936 ± 0.0062

0.8081 ± 0.0038

0.8560 ± 0.0162

0.7956 ± 0.0077

0.8016 ± 0.0190

Dataset 4

0.7585 ± 0.0119

0.6629 ± 0.0190

0.8032 ± 0.0104

0.6683 ± 0.0061

0.7261 ± 0.0145

0.8488 ± 0.0175

Dataset 5

0.8698 ± 0.0032

0.7943 ± 0.0019

0.8731 ± 0.0016

0.9596 ± 0.0021

0.8792 ± 0.0031

0.9457 ± 0.0033

Ave.

0.8334

0.8067

0.8620

0.8617

0.8041

0.8849