Skip to main content

Table 3 Performance evolution of poly(A) tail prediction models (Se sensitivity, Sp specificity, Acc accuracy)

From: From shallow to deep: some lessons learned from application of machine learning for recognition of functional genomic elements in human genome

Tool

Reference

Year

Adjusted values

Se

Sp

Acc

Polyadq

[36]

1999

46

86

65

PolyA Signal Miner

[37]

2003

72

80

 

ERPIN

[38]

2003

66

88

75

PolyA_SVM

[39]

2006

56

78

68

PolyFd/PolyFud

[40]

2009

72

80

78

Polyapred

[41]

2009

57

86

 

Polyar

[42]

2010

57

50

53

Chang et al

[43]

2011

56

90

75

DPS-ANN

[12]

2012

  

78

HMM-SVM

[44]

2013

80

87

81

DSET

[45]

2015

86

86

86

Omni_PolyA

[35]

2018

  

80

DeepGSR

[22]

2019

  

84

DeeReCT-PolyA

[46]

2019

  

84

  1. Entries with no value are explained in “Methods” section