/watsite_flexibility_scripts_data

Primary LanguagePython

watsite_flexibility_scripts_data

Step 1: run a long free MD simulation for the system (with or without ligand)

  • for protein-ligand complex, use the openmm_prep.py to define the interactions to be restrained
  • cluster the trajectory based on BindingSite rmsd
    • pymol -c getBindingSiteResi.pml will generate bs_res.txt, and modify cpptraj_*.in files with correct binding site residue numbers
    • align trajectory: cpptraj < 1_cpptraj_traj.in
    • cluster trajectory: cpptraj < 2_cpptraj_cluster.in --> cluster_hier.c? will be generated for the clusters
  • redo clustering for the top 10 clusters using the runCpptraj.py script python runCpptraj.py -c --> generate rep_c?.c?.pdb
  • plot 2drmsd.gnu for the top 10 clusters python runCpptraj.py -r --> generate rms2d_10cluster.gnu
  • For each cluster of interest, extract the conformation based on the frame number python runCpptraj.py -f hier.4_summary_sieve10.txt c0_summary.txt python runCpptraj.py -f hier.4_summary_sieve10.txt c1_summary.txt python runCpptraj.py -f hier.4_summary_sieve10.txt c4_summary.txt

Step 2: rerun short (5ns) WATsite for each subcluster from cluster

  • cluster folder created from extraction of previous long trajectory
  • run WATsite docker, and run python in the folder contains all cluster folder
python subAll.py

Analysis Scripts and Brief Usage

  • to generate an rms2d plot and RMSF values for binding site residues throughout a simulation, run cpptraj < cpptraj_rmsf.in
  • analyze hydration site predictions from all subclusters by running python analyzeAllHS.py Produces graph with HS sites identified as pairs and generates by_[0-10].pdb files
  • for grid analysis, use combineDX.py to generate mean density, standard deviation, and max error dx grids
  • to test the convergence of a system, use egy_convergence.py to compare the energies from the first frames to the entire trajectory. This script can also compare energies between different simulations (restrained vs unrestrained)
  • plotHist.py can also be used to test the positional convergence of hydration site predictions and to compare against crystal waters
  • plotPerHSenthalpy.py script can be used to vizualize the enthalpy values of each HS through out the trajectory. Useful when there is a large range of enthalpy.
  • to easily generate binding site crystal waters to compare against, issue the command python findCW.py -g grid_occupancy.dx -c crystal_structure.pdb

Collected Data for Publication

HSP90
  • 10clusters_rms2d.png RMS of BS residues between top 10 clusters after clustering
Restrained RMSD
  • k[0.6-4.8]rms2d.png RMS of BS residues throughout restrained simulations
Cluster Subcluster
  • FigureC[0,1,7].png Clustered hydration sites using ten subclusters
Truncated
  • thermo_prop_15A_15A4ns.png Energy property comparison between 10ns and 4ns 15A truncation
  • thermo_prop_15A_full.png Energy property comparison between 15A truncation and full protein
  • thermo_prop_20A_full.png Energy property comparison between 20A truncation and full protein
Klebe
  • k[1.2-5]rms2d.png RMS of BS residues throughout restrained simulations

Example