Skip to main content

Table 8 The \(K_{\mathbf{c}{} \mathbf{c}^\prime }\) matrix for kinase inhibitor generation task, the diagnal elements \(K_{\mathbf{c}{} \mathbf{c}}=R_\mathbf{c}\) are omitted since they have been reported in Table 7

From: Multi-objective de novo drug design with conditional graph generative model

Condition (\({\mathbf{c}}\))

Model

Results(\({\mathbf{c}}^\prime\))

GSK-3\(\beta\)(+),

JNK3(+)

GSK-3\(\beta\)(+),

JNK3(−)

GSK-3\(\beta\)(−),

JNK3(+)

GSK-3\(\beta\)(+)

Graph

–

\(0.178\pm 0.007\)

\({\it 0.018}\pm {\it 0.001}\)

JNK3(+)

SMILES

–

\({\it 0.167}\pm {\it 0.010}\)*

\(0.063\pm 0.006\)

GSK-3\(\beta\)(+)

Graph

\({\it 0.034}\pm {\it 0.001}\)***

–

\({\it 0.003}\pm {\it 0.000}\)***

JNK3(−)

SMILES

\(0.082\pm 0.007\)

–

\(0.023\pm 0.002\)

GSK-3\(\beta\)(−)

Graph

\({\it 0.024}\pm {\it 0.004}\)***

\({\it 0.022}\pm {\it 0.002}\)***

–

JNK3(+)

SMILES

\(0.083\pm 0.007\)

\(0.057\pm 0.002\)

–

  1. Results are reported as \(Mean \pm SD\). The best performance in each metric is highlighted in italics face.Paired t-tests are carried out for the difference between the graph and SMILES based method (*** for \(p\le 0.001\), ** for \(p\le 0.01\) and * for \(p\le 0.05\))