RI-Score
RI-Score/R
Use RI-Score with follwing multiscale-kernel parameters:
Lorentz kernel: tau=6.0, alpha=10.0, cutoff=37
Exponential kernel 1: kappa=3.5, alpha=1.0, cutoff=8
Exponential kernel 2: kappa=10, alpha=2.5, cutoff=17
Use RandomForest package in R with number of trees = 550, using a total of number of features
For each target ligand, use relibase server to collect all complexes having a similiarity ligand >= 0.3 and form a training set. RI-Score will learn that training data and predict the binding energy of the target ligand.
Maestro/Gold/RI-Score
Maestro Schrodinger Suite_2015-2/CCDC Gold Suite 2016/RI-Score
(prepwizard) -propka_pH # Use pH values listed in Data_set_fxr_crystallization_conditions.csv
(prepwizard) -fillsidechains -s # fillsidechains for target protein
(ligprep) -adjust_itc -ph # pH given in Data_set_fxr_crystallization_conditions.csv
Maestro's prepwizard was used to optimize the protein with pH value and fillsidechains option.
Maestro's ligprep was used to generate optimized 3d structure of ligands from 2d structure. Ligands without knowing the native 3D structure are docked into the closet protein (which binds to the most similar ligand) among the 36 ones provided by D3R.
autoscale = 0.2
floodfill_center = cavity_from_ligand 8 atoms
gold_fitfunc_path = plp # use ChemPLP as scoring function to generate poses
num_poses = 500 # generate 500 poses as pose_pool for each complex
Use Gold to dock ligand to the closet protein to generate 500 poses. After getting the pose_pool, use in house machine learning based score to rescore the pose_pool and choose a pose having the lowest score.