Skip to main content

Table 2 Performances of fivefold cross-validation for 18 targets using SVM

From: ExCAPE-DB: an integrated large scale dataset facilitating Big Data analysis in chemogenomics

Target Active compounds Inactive compounds Ratio (active/inactive compounds) Sensitivity Precision Specificity κ
PPARA 1955 1465 1.33 0.96 0.94 0.92 0.89
MMP2 2742 2363 1.16 0.96 0.96 0.96 0.92
MAOA 732 733 1.00 0.79 0.80 0.81 0.59
NR1I2 249 1090 0.23 0.82 0.73 0.93 0.72
TMPRSS15 139 724 0.19 0.43 0.54 0.93 0.39
HSD17B10 3410 11,510 0.30 0.41 0.40 0.82 0.23
KDM4E 3938 35,059 0.11 0.22 0.29 0.94 0.18
LMNA 14,533 171,164 0.09 0.49 0.13 0.72 0.10
TDP1 23,133 276,782 0.08 0.76 0.38 0.90 0.45
TARDBP 12,193 387,934 0.03 0.22 0.08 0.92 0.08
ALOX15 1932 69,362 0.03 0.49 0.12 0.90 0.16
BRCA1 8619 363,912 0.02 0.72 0.20 0.93 0.29
DRD2 4613 343,076 0.01 0.96 0.93 1.00 0.94
GSK3B 3334 300,186 0.01 0.85 0.72 1.00 0.78
JAK2 2158 213,915 0.01 0.85 0.81 1.00 0.83
POLK 773 389,418 0.002 0.55 0.17 0.99 0.26
FEN1 1050 381,575 0.003 0.35 0.03 0.96 0.04
HDAC3 369 311,425 0.001 0.98 0.76 1.00 0.86