Lead Finder + reduced xedex conformer sampling
Lead Finder 1808, xedex, Build Model 1808
Ligands 3D optimized using xedmin tool
Rings conformations generated using xedex, (-D 50 -m 20)
Build Model 1808 with default settings was used for preparation of protein models.
Ligands were converted to 2D using babel, 3D optimized using xedmin tool. Ligand macrocycle conformations were
sampled using xedex tool (with -D 50 -m 20 option). All aliphatic NH were manually protonated to (+) state,
amide bonds in cyclic rings were set to trans- conformation. 3 protein models were used for docking
10 independent docking runs / model + ligand + conformation
default docking parameters
10 independent docking runs of Lead Finder 1808 with default parameters were used per
conformation per model. H-bonds filters were applied to poses: h-bonds with residue 230, and (residue 32
or residue 228. Filtered poses were sorted by Lead Finder dG, and poses within 1 kcal/mol of minimum were
inspected and rated (A = best, B = some H-bonds missing, C = many H-bonds missing, D = incorrect). Pose with
the lowest dG of the highest rate were selected for each ligand. Ligands, for which no poses satisfying
hbonds and dg filters were reviewed and rated manually, and same procedure was applied.
Yes
Lead Finder dG score plus
Lead Finder 1808
default Lead Finder parameters
Standard Lead Finder dGs were calculated for poses and poses were manually rated into 3 categories by
the quality of the best found pose. Score was calculated as dG + rank*1.7, where 1.7 is 2 standard deviations
of dG for top rank poses, rank = 0, 1 and 2.
Yes