TI and strain calculations

Protocol for Thermodynamic Integration using Amber 18

1. Selection of the binding site for docking:
   1. Identify the binding site for search of the protein where the ligand is going to be docked.
   2. Define a search box around the selected site and run the docking software selected for the search.
2. Preparation of the protein-ligand complex:
   1. Extract the protein and ligand coordinates from the best docking pose file.
   2. Prepare the protein and ligand structures for molecular dynamics simulations by adding missing atoms, assigning protonation states, and neutralizing the system with counterions if necessary.
   3. Assign force field parameters to the protein and ligand molecules.
   4. Add water molecules to solvate the protein-ligand complex and create a periodic simulation box. The size of the box should be large enough to avoid artificial interactions between periodic images of the system.
3. Preparation of the simulation system:
   1. Minimize the energy of the system using an energy minimization algorithm to remove any steric clashes or bad contacts.
   2. Equilibrate the system using a suitable equilibration protocol (e.g. NVT or NPT) to allow the water and ions to equilibrate around the protein-ligand complex.
4. Molecular dynamics simulations:
   1. Run several production molecular dynamics simulations of the protein-ligand complex in explicit solvent in NPT, with different initial velocities or random seed values.
   2. For each simulation, save the coordinates and velocities of the system at regular intervals for later analysis.
5. Analysis from unbiased trajectories:
   1. From the saved trajectories in the previous step, discard the detached ligands from the protein.
   2. From the stable ligands, apply a clustering algorithm on the trajectory to find the stable poses in the simulation and use them for the free energy calculation.
6. Calculation of the free energy:
   1. In this protocol Thermodynamic Integration (TI) is used to calculate the absolute binding free energy of the ligand to the protein. This involves performing additional molecular dynamics simulations to compute the free energy changes associated with the ligand transformation from the bound state to the unbound state.
   2. From the selected final poses obtained in the analysis of the trajectory, initialize the system using those ones that also will serve as reference for the restraint used in calculations.
   3. In the amber18 manual it is described the keywords used in the pmemd input file to activate the TI calculation, the most important part is to correctly select the residue number that matches the ligand in the topology file. Since we are using total annihilation of the residue, a second topology file is not required. Also activate the parameters described in the manual for the one step annihilation of the residue and include the positional restraint of the ligand.
   4. After selecting the number of windows per TI calculation, run three independent complete replicas for each residue changing the initial velocities.
7. Analysis and validation of the results:
   1. Analyze the binding energy of the ligand obtained from the simulations. The amber manual web page have links to tutorials for TI calculations that include example scripts to extract the information to calculate the dH/dλ for each window and perform the trapezoidal integration to calculate the absolute binding free energy for each replica.
   2. Calculate the free energy correction for the restraint using the equation shown in the article from this protocol, where the value of the restrain constant is required. 

Protocol for strain analysis of protein simulations.

1. Prepare the reference and system of interest.
   1. Follow the steps 1-4 of the previous section for the reference system.
   2. Follow steps 1-5 of the previous section for the system of interest.
2. Run the strain analysis.
   1. Use the attached python script with the saved trajectories of the previous step.