understand MCCE
We will try to explain how MCCE does its calculations through the following 3 residues from (4LZT) pdb.
The following are the x y z coordinates from the X-RAY DIFFRACTION, res # 5, 18 and 29.
*Image is taken in pymol
Step 1: Preparing the Protein
Residue topology files for each amino acid and ligand define the heavy atom bond connectivity, the number and position of hydrogens to be added to each atom, rotamer building rules.
In this step, mcce calculates the solvent accessible surface (SAS) area and strips off exposed water and salt (HOH, NO3 and SO4).
Step 2: Making rotamers
The protein is divided into fixed backbone and flexible side chains. MCCE needs to produce an ensemble of low energy side chain positions to allow the protein to remain in equilibrium with the different ionization states found for example in a pH titration. Step 2 has 8 stages.
Stage a: Protein Side Chain Optimization and Relaxation
A set of ideal rotamers is created with ideal bond lengths, bond angles, and dihedral angles. The heavy atom rotamer closest to that found in the crystal structure is kept. Then all ideal rotamers for the protein are minimized using the steepest decent method81,82 with Amber nonelectrostatic parameters, PARSE charges and a uniform dielectric constant of 6, assuming standard ionization states with His neutral.
Step 3: Calculating the energy lookup table
In this step, mcce calculates four energy functions:
Energy look-up table is prepared, allowing Monte Carlo sampling for all microstate energies. Thus, for M conformers, there are four M dimensional vectors containing terms assumed to be independent of the selected conformers for other residues. There are two symmetric MxM matrices for the conformer-conformer electrostatic and the LJ interactions. The electrostatic interactions are calculated with the Poisson-Boltzmann equation using multiple DelPhi runs integrated in MCCE.