Skip to main content

Advertisement

Table 11 Mean Δ EF values and standard deviations for the MCS and MCS ext similarity methods (approach B2)

From: Improving structural similarity based virtual screening using background knowledge

DuD set MCS MCS ext
  1% 5% 10% 1% 5% 10%
HMGR 8.5 ± 4.5 7.0 ± 6.8 2.8 ± 3.6 9.1 ± 2.5 6.5 ± 6.1 2.0 ± 2.8
ER 13.6 ± 7.6 12.6 ± 4.1 5.3 ± 1.7 10.2 ± 6.5 10.6 ± 2.5 5.0 ± 1.0
PPAR γ 4.6 ± 10.6 1.2 ± 5.4 1.7 ± 2.8 4.5 ± 11.0 3.8 ± 5.5 1.6 ± 2.9
P38 MAP 9.6 ± 7.9 8.6 ± 3.7 3.3 ± 1.8 7.1 ± 6.8 7.3 ± 3.7 2.7 ± 2.0
TK 20.1 ± 4.4 12.6 ± 2.1 5.1 ± 1.6 19.7 ± 5.3 14.0 ± 2.1 5.5 ± 1.5
FXa 4.6 ± 11.2 7.6 ± 3.8 3.3 ± 1.8 3.5 ± 11.2 6.2 ± 4.7 2.5 ± 2.6
ADA 10.1 ± 6.4 8.2 ± 3.0 4.3 ± 3.6 12.8 ± 6.4 8.8 ± 3.3 6.1 ± 4.6
DHFR 10.9 ± 10.6 11.7 ± 2.9 4.7 ± 1.1 3.4 ± 7.0 2.4 ± 2.1 0.1 ± 0.1
AChE 10.3 ± 12.5 11.3 ± 4.7 4.8 ± 2.5 10.1 ± 11.9 10.4 ± 5.1 4.4 ± 3.0
COX-2 12.3 ± 9.2 11.7 ± 2.2 5.3 ± 1.1 11.4 ± 10.3 10.5 ± 3.7 2.5 ± 2.5
w/d/l   8 / 0 / 2 7 / 0 / 3 8 / 0 / 2
  1. Mean Δ EF and standard deviations for the MCS and MCS ext similarity methods at 1%, 5% and 10% of the database (receptor specific decoy set DuD set ). The extension fingerprint is calculated from 10% (20% for HMGR, TK and ADA) of the ligands (approach B2). Improvements of MCS ext compared to MCS are marked with bold print. w/d/l = wins/draws/losses.