Name

In_house_machine_learning_score/Maestro/GOLD

Software

Maestro_Schrodinger Suite_2016-2/CCDC_Gold Suite_2016/R

Parameters

Use PH values listed in Data_set_fxr_crystallization_conditions.csv, use other default parameters in Ligprep of Schrodinger
Gold suite with autoscale = 0.2 and goldscore as a scoring function
RandomForest with number of trees = 550, using a total of number of features

Method

Method: The training set is from PDBbind refined set v2015. Gold docking software is used to get decoy sets, and the top poses are selected by our in-house software. All features of the training and test sets are generated by our in house machine learning tool. RandomForest is used to get predicted values.

Name

Maestro/Gold/FRI-Score

Software

Maestro Schrodinger Suite_2016-2/CCDC Gold Suite 2016

System Preparation Parameters

(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

System Preparation Method

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. Sample ligand was manually filled into the binding pocket as the reference ligand. Position of the reference ligand was set as binding site.

Pose Prediction Parameters

autoscale = 0.2
floodfill_center = cavity_from_ligand 8 atoms
gold_fitfunc_path = goldscore # use goldscore as scoring function to generate poses
num_poses = 500 # generate 500 poses as pose_pool for each complex

Pose Prediction Method

Use Gold to get top 50 poses. The top 5 candidates are selected by our in house machine learning.