Skip to main content

Table 3 NDCG@10 of strategy I

From: When drug discovery meets web search: Learning to Rank for ligand-based virtual screening

  AdaRank RankNet ListNet PRank RankBoost SVMRank SVR
ADORA3 0.4463 0.5885 0.5119 0.4032 0.6543 0.6446 0.6815
BDKRB2 0.5549 0.6186 0.4208 0.6242 0.5564 0.5917 0.5935
CB1 0.5913 0.4983 0.4586 0.6052 0.6993 0.7026 0.6921
CTSK 0.4225 0.3850 0.4741 0.4673 0.6545 0.5253 0.5199
CCK1 0.4122 0.5110 0.5704 0.7661 0.8523 0.7673 0.7136
CHRM1 0.1254 0.2978 0.1825 0.5366 0.6341 0.7076 0.7068
CHRM3 0.3295 0.5366 0.4880 0.7282 0.9019 0.8738 0.7277
TOP1 0.2076 0.3441 0.5005 0.6284 0.7746 0.8101 0.7387
EPHX2 0.4749 0.5997 0.5481 0.5506 0.6604 0.6102 0.5913
FBPase 0.5476 0.5420 0.5328 0.6281 0.8081 0.7810 0.7710
HMGCR 0.4078 0.5584 0.5475 0.6169 0.8089 0.7956 0.7660
Itgαvβ3 0.4168 0.3436 0.3555 0.4605 0.5837 0.5399 0.5360
JAK2 0.4208 0.3270 0.4184 0.5256 0.6804 0.6653 0.6548
KIF11 0.4682 0.4684 0.5724 0.5172 0.6912 0.7267 0.7163
LXR-beta 0.5828 0.5293 0.5009 0.6288 0.7260 0.7104 0.6899
mTOR 0.5204 0.5169 0.4038 0.6657 0.8334 0.8357 0.8517
MK2 0.5860 0.4398 0.4510 0.5909 0.7299 0.6945 0.7272
MMP-8 0.5792 0.4819 0.4843 0.5758 0.6699 0.6841 0.6815
ORL1 0.6082 0.3600 0.6024 0.6530 0.7270 0.7656 0.7430
PDE5 0.4877 0.6042 0.4628 0.5718 0.7368 0.7237 0.7117
EP3 0.4489 0.4484 0.5028 0.4054 0.6504 0.6292 0.6306
SGLT2 0.4619 0.3547 0.4285 0.5053 0.7047 0.6826 0.6843
CYP17 0.4829 0.4057 0.4001 0.4823 0.5637 0.4887 0.5231
ASC 0.4251 0.3584 0.4199 0.5630 0.6243 0.5629 0.5813
  1. The bold number among each row indicates the best performance among all the methods in this row.