Allosteric Site Searching with Conditional Mutual Information

Requirements:

• CHARMM c48b2
• PyMol 2.5+
• matplotlib
• mdtraj==1.9.9
• numpy
• sklearn

Step 1: Generate Inputs

Use CHARMM-GUI PDB reader & manipulator to generate the input files. https://www.charmm-gui.org/?doc=input/pdbreader

We used 6DG8 from RCSB and translated its center of mass to the origin with the following script using PyMol.

python code\align\align.py
python code\align\crd.py

Step 2: Run CHARMM Monte Carlo

Generate the monte carlo ensembles for all six ligand categories:

bash code/monte_carlo/run_MC.sh

This will run the six monte carlo simulations in parallel.

Step 3: Obtain Observables

Check the energy autocorrelation function:

charmm -i code/energy_acf/acf.inp
python code/energy_acf/acfPlot.py

the equilibrated frame is the start frame for the next commands.

Calculate the Ion Channel Barrier of each frame:

bash code/channel_barrier/calc_barrier.sh

Extract the energy of amino acids:

for n in 0 1a 1b 1c 1d 1e 2ab 2ac 2ad 2ae 2bc 2bd 2be 2cd 2ce 2de 3abc 3abd 3abe 3acd 3ace 3ade 3bcd 3bce 3cde 4abcd 4abce 4abde 4acde 4bcde 5
do
charmm -i code/nrgs/nrgs$n.inp
done
python code/nrgs/read.py ### at this step, still need to fix

Step 4: Calculate Thermodynamic Coupling Function

python code/TCF.py

-- Visualization --

Open the input/6dg8_aligned.pdb with PyMol, paste the contents of output/i3colors.txt then output/i3set.txt directly into PyMol.

-- Deprecated --

Extract the position of amino acids:

python code/positions/extractPositions.py

Calculate Conditional Mutual Information

python code/I3.py