Skip to main content

Table 4 Performance of the pharmacophore models based on hiphop algorithms

From: Discovery of novel JAK1 inhibitors through combining machine learning, structure-based pharmacophore modeling and bio-evaluation

No

Feature

Ranking score

Direct hit (DH)

Partial hit (PH)

Max fit

Cluster

Cutoff

Precision

Recall

F1 score

Mcc

Hiphop1

RHHDAA

91.652

111111

000000

6

I

2.3290

0.7839

0.06338

0.1173

0.1511

Hiphop2

RHHDAA

91.151

111111

000000

6

I

     

Hiphop3

HHDDAA

90.903

111111

000000

6

II

2.2359

0.9020

0.1080

0.1929

0.2377

Hiphop4

HHDDAA

86.869

111111

000000

6

II

     

Hiphop5

RHDAA

85.124

111111

000000

5

III

2.5917

0.6407

0.08790

0.1546

0.1305

Hiphop6

RHDAA

84.703

111111

000000

5

III

     

Hiphop7

RHDAA

83.264

111111

000000

5

III

     

Hiphop8

RHDAA

83.264

111111

000000

5

III

     

Hiphop9

RHDAA

83.264

111111

000000

5

III

     

Hiphop10

RHDAA

83.030

111111

000000

5

III

     
  1. The bold indicates the optimal model of different Hiphop models