Skip to main content

Table 1 ROC scores of one transcriptomics and two proteomics datasets

From: A two-step site and mRNA-level model for predicting microRNA targets

Dataset

 

SVM

PITA

TScan

miRan

PITAT

TS_C

mirT2

PicTa

Linsley

All

0.81

0.76

0.75

0.55

0.60

0.61

0.63

0.58

 

ROC10*n

0.0129

0.0018

0.0173

0.0105

0.0035

0.0170

0.0196

0.0113

 

7m+C

0.73

0.61

0.69

0.43

0.57

0.59

0.67

0.57

Selback

All

0.64

0.61

0.61

0.52

0.55

0.55

0.54

0.53

 

ROC10*n

0.0253

0.0042

0.0212

0.0079

0.0138

0.0213

0.0231

0.0210

 

7m+C

0.71

0.61

0.69

0.42

0.60

0.63

0.60

0.58

Baek

All

0.56

0.56

0.56

0.51

0.52

0.53

0.52

0.52

 

ROC10*n

0.0193

0.0046

0.0157

0.0081

0.0148

0.0174

0.0086

0.0131

 

7m+C

0.59

0.60

0.62

0.44

0.56

0.61

0.54

0.54

  1. Three benchmarks, All, ROC10*n, 7m+C (7mer+Conservation), were performed on one transcripomics (Linsley) and two proteomics (Baek and Selback) datasets. The ROC scores were calculated for eight algorithms, SVM, PITA, TScan (TargetScan), miRan (miRanda), PITAT (PITA Top), TS_C (TargetScan with conserved genes), mirT2 (mirTarget2) and PicTa (PicTar)