Skip to main content

Table 4 Comprehensive assessment of PUResNetV2.0 and GRaSP on benchmark datasets

From: PUResNetV2.0: a deep learning model leveraging sparse representation for improved ligand binding site prediction

Benchmark dataset

Methods

Precision (%)

Recall (%)

F1 score (%)

MCC (%)

Coach100

PUResNetV2.0

61.3

62.4

61.8

62.4

GRaSP

63.5

43.9

51.9

50.1

Holo801

PUResNetV2.0

70.2

71.0

70.6

68.3

GRaSP

66.9

50.1

57.3

55.1

Apoholo45

PUResNetV2.0

69.7

54.2

61.0

59.4

GRaSP

61.7

32.0

42.1

42.0

PDBBind1681

PUResNetV2.0

78.68

35.90

47.10

46.70

GRaSP

57.31

22.89

31.01

30.84

  1. Top values for each benchmark dataset are represented in bold