How to generate a free energy Ramachandran plot from a GROMACS MD trajectory if protein has more than 2 residues? : comp_chem

I ran some REMD calculations with GROMACS and wish to create a plot which depicts the free energies of the different regions on the Ramachandran plot of my protein at a specific temperature, similar to figure 2 in the following paper below.

Article Antifreeze Glycoproteins Bind Reversibly to Ice via Hydrophobic Groups

This seems somewhat straightforward to do when the peptide only has 2 residues, meaning there is only a single pair of dihedral angles. However, the protein they studied has 14 residues, so I am not too sure how they were able to create a 3-dimensional free energy plot when there’s 28 dihedral angles and the dimensional space is much greater than 3. Unfortunately, the paper did not include much details about how they computed figure 2.

I also wish to use the Weighted Histogram Analysis Method (WHAM) in order to use data from every replica and temperature. However, many implementations of WHAM that I found for creating free-energy Ramachandran plots (e.g. ronlevygroup.cst.temple.edu/software/UWHAM_and_SWHAM_webpage/03_AlaD_Replica_Exchange/readme.html#orgd0f52e4 or www.strodel.info/index_files/lecture/html/enhsampling_remd.html ) seems to only work for two-residue peptides, like dialanine, as their input files seem to require matching each pair of phi/psi angle at each timestep with the associated potential energy at that timestep. Again, I am not sure how this is possible when the protein has 28 dihedral angles and thus a much larger dimensional space.

Read more here: Source link