Skip to main content

Table 2 The second and third columns show the average and standard deviation of AUCs on 585 diseases and 137 drugs for each subset of drug features, respectively

From: The assessment of efficient representation of drug features using deep learning for drug repositioning

Subset E

Avg ∓ STDV of AUC on 585 disease

Avg ∓ STDVof AUC on 137 drugs

{s}

0.909 ∓ 0.08

0.802 ∓ 0.14

{g}

0.724 ∓ 0.18

0.837 ∓ 0.10

{e}

0.495 ∓ 0.19

0.921 ∓ 0.09

{p}

0.620 ∓ 0.22

0.939 ∓ 0.05

{g,s}

0.911 ∓ 0.08

0.790 ∓ 0.15

{e,s}

0.821 ∓ 0.11

0.795 ∓ 0.15

{s,p}

0.896 ∓ 0.09

0.807 ∓ 0.14

{e,g}

0.644 ∓ 0.20

0.839 ∓ 0.11

{g,p}

0.713 ∓ 0.19

0.836 ∓ 0.11

{e,p}

0.570 ∓ 0.20

0.920 ∓ 0.06

{e,g,s}

0.797 ∓ 0.14

0.792 ∓ 0.15

{g,p,s}

0.833 ∓ 0.12

0.798 ∓ 0.14

{g,p,e}

0.687 ∓ 0.19

0.832 ∓ 0.12

{e,p,s}

0.822 ∓ 0.11

0.797 ∓ 0.14

{s,e,g,p}

0.798 ∓ 0.14

0.792 ∓ 0.15