Data for "Leveraging the Thermodynamics of Protein Conformations in Drug Discovery" submitted to JCIM
Input structures for ABFEP calculations.
Input structures for PReorg calculations. These input structures were chosen from MD simulations. Starting from each PDB structure, we ran three 75ns long MD simulations. Then the structures generated by MD simulations were clustered by comparing the conformation of the alchemical part. If the initial PDB structure is in the largest cluster, the PDB structure was chosen as the input structure for the PReorg calculations. If the PDB structure is not in the largest cluster, the centroid of the largest cluster was chosen as the input structure for the PReorg calculations.
Raw data for Fig.2 and Fig.3.
Python script used to cluster the structures generated by MD simulations. Example:
$SCHRODINGER/run cartesian_custom_dist_clustering_trajectory_json_prop_iterator.py \
PDB.mae traj01.cms traj02.cms traj03.cms \
--reference PDB.mae \
--fitasl "ASL to select the template part" \
--asl "ASL to select the alchemical part" \
--threshold 4 \
--softness 0.5 \
--cutoff 4 \
--min_size 5 \
--extract best \
--num_reps 1 \
-j JOBNAME