Simulate liposome with hundreds to thousands lipids require a large amount of resources to simulate for all-atom model, which is almost impossible for most researchers, CG model is a techniques allow considerable extension of accessible size and time in simulations of biological systems. although many CG model have been reported for proteins and DNAs, very few have been for lipids, the object of this study is to develop a new pH sensitive CG model for lipids, here we use MORC16 as an example. We refer to MARTINI and SIRAH to help us model new designed lipids.

 

Step 1. Mapping Atoms to Coarse Grains

To map atoms to coarse grained model, we use cgconv.pl in SIRAH package, the only thing need to do is define the map strategy, we saved it as sirah_lipid.map

/home/zwei/work/MD/sirah.amber/tools/CGCONV/maps/

/home/zwei/work/MD/sirah.amber/tools/CGCONV/cgconv.pl -i morc16.pdb -o morc16_cg.pdb

Figure 1. Mapping scheme between atomistic and CG model of MORC16. Colored spheres are elements from all-atom representation preserved in CG model.The strategy for coarse graining is to keep key atoms which have more chmical sense while ignore other atoms which only act as structural supporting atoms. Here we keep all non-carbon heavy atoms and delete some carbon atoms on six-member ring and all hydrogen atoms.

Step 2. Creating Parameter Files

We build library file and parameter file by referring other lipid models. See Amber14 manual 13.1.4. Parameters required for amber force fields.

Download morc16_cg.frcmod, morc16_cg.lib ,leaprc.cgla to current directory,

#load lipid_go forcefield                                                               source leaprc.cgla #load model mor=loadpdb morc16_cg.pdb #Save Parms saveamberparm mor morc16_cg.prmtop morc16_cg.inpcrd #EXIT quit

tleap -f tleap.in

Minimization &cntrl   imin = 1,   ncyc=10000,   maxcyc = 20000,   ntpr=100,   ntb=0,   igb=0,   cut=12 /

Heating and Equilibrating  &cntrl    imin=0,irest=0, ntx=1,ntxo=1,    ntpr=1000, ntwx=1000, nstlim=30000,    dt=0.002, ntt=3, tempi=50.0,nscm=100,    temp0=300, gamma_ln=1.0, ig=-1,    ntb=0,igb=0,ntc=2, ntf=2, cut=12,                                                                  ioutfm=1,ntrelax=100,  / &wt type='TEMP0', istep=50.0,istep2=5000,value1=0.0,value2=300.0 / &wt type='TEMP0', istep=5001,istep2=30000,value1=300.0,value2=300.0 / &wt type='END' /

Production Run  &cntrl    imin=0, irest=0, ntx=1,ntxo=1,    ntpr=10000, ntwx=10000,ntpr=10000,nstlim=1000000,    dt=0.002, ntt=3,nscm=100,    temp0=300, gamma_ln=1.0, ig=-1,    ntb=0,igb=0,ntc=2, ntf=2, cut=12,    ioutfm=1,ntrelax=100,  /

sander -O -i min.mdin -o morc16_cg_min.out -p morc16_cg.prmtop  -c morc16_cg.inpcrd -r morc16_cg_min.rst

sander -O -i heat.mdin -o morc16_cg_heat.out -p morc16_cg.prmtop -c morc16_cg_min.rst -r morc16_cg_heat.rst

sander -O -i md.mdin -o morc16_cg_md.out -p morc16_cg.prmtop -c morc16_cg_heat.rst -r morc16_cg_md.rst -x morc16_cg_md.mdcrd

***pmemd does not explicitly support vacuum simulations

 

Step 3. Micelle Modelling

To compare with all-atom micelle simulation, we build initial coarse grained structure from all-atom structure.

/home/zwei/work/MD/sirah.amber/tools/CGCONV/cgconv.pl -i morc16_micelle_packmol.pdb -o morc16_cg_micelle_packmol.pdb

#load lipid_cg forcefield source /home/zwei/work/MD/MORC16/CGLA/leaprc.cgla #load model mor=loadpdb morc16_cg_micelle_packmol.pdb #add solvent and ions solvatebox mor WT4BOX 40                 #Save Parms saveamberparm mor morc16_cg_micelle.prmtop morc16_cg_micelle.inpcrd #EXIT quit

# # ----- leaprc for loading the CGLA lipid forcefield # # CGLA atom types # Setting sp3 prevents froms confusing messages about hybridization    addAtomTypes { # MOC ------------------------------                { "bc" "" "sp3" }                { "bn" "" "sp3" }                { "bo" "" "sp3" }                { "do" "" "sp3" }                { "lc" "" "sp3" }                { "co" "" "sp3" }                { "lo" "" "sp3" }                { "gc" "" "sp3" }                { "tc" "" "sp3" }                                                                                                                                                                                  # Solvent --------------------------                { "WT" "" "sp3" }                { "KW" "" "sp3" }                { "NaW" "" "sp3" }                { "ClW" "" "sp3" }   } # Load SIRAH parameters   loadAmberParams morc16_cg.frcmod   loadAmberParams solv.frcmod # Load SIRAH libraries   loadOff morc16_cg.lib   loadOff solv.lib   loadOff wt4box.off

tleap.in

leaprc.cgla

 

tleap -f tleap.in

Minimization of water box               &cntrl    imin = 1,    ncyc=10000,     maxcyc = 20000,     cut = 8,     ntpr=100,   /

heating and equilibrating  &cntrl    imin=0, irest=0, ntx=1,ntxo=1,    ntpr=1000, ntwx=1000, nstlim=30000,    dt=0.002, ntt=3, tempi=50.0, ntp=1,    temp0=300, gamma_ln=5.0, ig=-1,    ntc=2, ntf=2, cut=8, iwrap=1,    ioutfm=1,ntrelax=100,  / &wt type='TEMP0', istep=50.0,istep2=5000,value1=0.0,value2=300.0 / &wt type='TEMP0', istep=5001,istep2=30000,value1=300.0,value2=300.0 / &wt type='END' /

Explicit solvent production run  &cntrl    imin=0, irest=1, ntx=5,ntxo=1,    ntpr=10000, ntwx=10000,ntpr=10000,nstlim=1000000,    dt=0.002, ntt=3,ntp=1,    temp0=300, gamma_ln=5.0, ig=-1,    ntc=2, ntf=2, cut=8,iwrap=1,    ioutfm=1,ntrelax=100,  /

mpirun -np 24 pmemd.MPI -O -i min.mdin -o morc16_cg_micelle_min.out -p morc16_cg_micelle.prmtop -c morc16_cg_micelle.inpcrd -r morc16_cg_micelle_min.rst

mpirun -np 24 pmemd.MPI -O -i heat.mdin -o morc16_cg_micelle_heat.out -p morc16_cg_micelle.prmtop -c morc16_cg_micelle_min.rst -r morc16_cg_micelle_heat.rst

mpirun -np 24 pmemd.MPI -O -i md.mdin -o morc16_cg_micelle_md.out -p morc16_cg_micelle.prmtop -c morc16_cg_micelle_heat.rst -r morc16_cg_micelle_md.rst -x morc16_cg_micelle_md.mdcrd

Step 4. Steered Molecular Dynamics of Micelle 

./create_SMDINPUTS.sh 10 1000000 morc16_cg_micelle.prmtop morc16_cg_micelle_md.rst

./create_job.sh 10 morc16_cg_micelle_md.rst 1

./job.1.sh