Skip to main content

Advertisement

Table 15 Mean Δ EF and standard deviation using the best α coefficients (approach B1)

From: Improving structural similarity based virtual screening using background knowledge

  MCS ext ECFP ext
DuD set 1% 5% 10% 1% 5% 10%
HMGR 5.8 ± 10.0 1.7 ± 0.7 0.6 ± 0.3 0.6 ± 1.9 2.6 ± 3.2 0.6 ± 0.1
ER 12.1 ± 6.0 9.3 ± 3.5 3.2 ± 1.1 6.0 ± 5.3 8.5 ± 2.1 3.1 ± 1.2
PPAR γ 4.5 ± 10.6 3.8 ± 5.5 1.5 ± 2.9 4.1 ± 10.7 3.6 ± 5.6 1.7 ± 2.5
P38 MAP 2.8 ± 6.9 4.8 ± 4.2 2.4 ± 2.1 2.7 ± 6.0 4.8 ± 4.2 2.4 ± 2.1
TK 18.3 ± 5.3 11.1 ± 3.8 3.7 ± 1.7 16.5 ± 8.4 11.1 ± 3.8 4.2 ± 1.9
FXa 3.5 ± 11.0 4.3 ± 5.4 2.0 ± 2.7 3.5 ± 11.0 4.2 ± 5.4 2.0 ± 2.7
ADA 9.2 ± 4.6 5.2 ± 0.0 2.2 ± 0.8 7.8 ± 7.5 5.2 ± 0.0 2.2 ± 0.7
DHFR 2.7 ± 5.1 0.0 ± 0.0 0.0 ± 0.0 1.9 ± 0.9 0.0 ± 0.0 0.0 ± 0.0
ACHE 10.0 ± 11.8 9.0 ± 6.0 4.3 ± 2.8 11.0 ± 12.0 9.0 ± 6.1 4.0 ± 2.8
COX-2 9.9 ± 9.8 9.8 ± 3.7 2.1 ± 2.6 6.7 ± 10.2 5.3 ± 5.0 2.1 ± 2.6
w/d/l 10 / 0 / 0 9 / 0 / 1 10 / 0 / 0 8 / 0 / 2 6 / 0 / 4 8 / 0 / 2
  1. Mean Δ EF and standard deviation using the best α coefficients for extended similarites MCS ext and ECFP ext for the receptor specific decoy sets DuD set at 1%, 5% and 10% of the database. The extension fingerprint is calculated from all ligands (approach B1). Improvements of MCS ext compared to MCS as well as ECFP ext compared to ECFP are marked in bold print. w/d/l = wins/draws/losses.