FEPrepare is NOT available as a stand alone program! To use FEPrepare please use FEPrepare web-server
FEP prepare automates the set-up procedure for performing NAMD/FEP simulations.
Install VMD
Align your ligands and protein
a) names3.py
ensures consistency between the atoms of the ligands and it creates the following files: newligandA.txt, newligandB.txt, newligandArtf.txt, newligandBrtf.txt, newligandAprm.txt, newligandBprm.txt, how_many.
b) merge2.py
identifies the common atoms between reference and mutant ligands. It returns sortedB, which is the .rtf file of the mutant ligand sorted according to the reference ligand.
c) dual2.py
implements the dual topology methodology. It generates files where both ligands coexist (pdb: hybridpdb.txt, rtf: final.txt, prm: updated.prm)
d) complex.py
generates a complex file where the hybrid ligand generated above and the protein coexist.
e) split_chains.py
generates the psfgen file needed from VMD.
f) vmd -dispdev text -e psfgen
g) if you would you like to add 150μΜ NaCl to your system:
vmd -dispdev text -e vmd_prepare_complex_after_gui_autopsf_ionize > vmd_log.txt
otherwise:
vmd -dispdev text -e vmd_prepare_complex_after_gui_autopsf > vmd_log.txt
h) vmd -dispdev text -e psfgen_solv
i) if you would you like to add 150μΜ NaCl to your system:
vmd -dispdev text -e vmd_prepare_ligand_after_gui_autopsf_ionize > vmd_log.txt
otherwise:
vmd -dispdev text -e vmd_prepare_ligand_after_gui_autopsf > vmd_log.txt
j) fep.py
generates the necessary ionized_fep files for the simulation.
k) min-max.py
returns the coordinates of the center of the box.