Skip to main content

Table 5 Frequently occurring substructures(PubChem fingerprint) of the drugs in NR

From: A unified solution for different scenarios of predicting drug-target interactions via triple matrix factorization

Substructures

Group of PubChem fingerprint

Occurrence (> = 75%)

'C-C-C-C-C-C-C'

G6: Simple SMARTS pattern

0.8462

'C-C-C-C-C-C-C-C'

G6: Simple SMARTS pattern

0.8077

'C(-C)(-C)(=C)'

G5: Detailed atom neighborhood

0.8077

'> = 16 H'

G1: Hierarchic Element Count

0.8077

'Cc1cc(O)ccc1'

G7: Complex SMARTS pattern

0.7692

'C-N-C-[#1]'

G6: Simple SMARTS pattern

0.7692

'C(~H)(~N)'

G4: Simple atom nearest neighbor

0.7692

'> = 16 C'

G1: Hierarchic Element Count

0.7692