Umbrella Sampling Simulations Of Tm Domain Dimerization

Umbrella sampling simulations of dimer formation of 2 TM domain dimers in DMPC/DLPC bilayers. Simulations are performed in the coarse-grained scheme using different force fields; normal Martini (N), Martini with all protein–protein interactions scaled (U) by 10% (U_10) or 20% (U_20), Martini with interactions among water-interacting beads scaled (W) either by 60% (W_60), 80% (W_80), or 90% (W_90), or the polarizable Martini (P). The original Martini is also repeated with an older set of 'common' simulation parameters (C).The tar files are named after the PDB codes of the corresponding dimer and the type of force field employed (see above). The tar files contain the run input files (.tpr) and the corresponding simulation parameter files (.mdp) for all umbrella windows; i.e. "EPHA_U_80_18.tpr" is the run input file for the EPHA dimer with all protein–protein interactions scaled down by 20% and with a protein–protein distance restrained to 18 Å by the umbrella potential. This tpr is generated from the simulation parameter file, the topology file (here EPHA_U_80.top), the index file (here EPHA.ndx), and the initial structure (here EPHA_start_18.gro, available in the EPHA-frames.tar). The index and initial structures for the polarizable model differ, and are provided in "EPHA-P.ndx" and "EPHA-frames-P.tar", respectively. All topologies (.itp) are provided in TOP.tar.The scaling is achieved by adding 'p' to the bead types in the proteins; either all types (scaling U) or to those more in contact with water than the membrane (scaling W) The corresponding parameters are given in the "martini_v2.2_scaled_X.itp" file. Note that for uniform style, the unscaled parameters are given in a similar manner in a file "martini_v2.2_unscaled.itp". Here, the .itp files follow the naming convention of the paper (see below) so that X=1 means downscaling of LJ epsilon by 10%, i.e. it corresponds to files with "_10".For a more thorough explanation of the purporse of the files and the simulation parameters, see the related publication:Javanainen M, Martinez-Seara H, Vattulainen I (2017) Excessive aggregation of membrane proteins in the Martini model. PLoS ONE 12(11): e0187936. https://doi.org/10.1371/journal.pone.0187936