Biomodeling: PMF of DNA-DNA Interactions

Preparation of simulation with a pair of dsDNA molecules

1. Prepare PDB and ITP files for a PBC dsDNA molecule.

   1. You can create new DNA moecules under PBC using our web server (http://yoo.skku.edu/apps)

   2. Otherwise, you can reuse PDB and ITP files from previous simulations.

      1. For example, I copied the following files from the simulation of DNA-PCNA complex.

         • Structure file: dna.pdb

         • Topology file for the system: topol.top

         • Moleculetype files for DNA: dna.itp dna.hbonds.itp  dna.pbc.itp

         • MDP files: grompp.mdp  equil.mdp mini.mdp

      2. Remove everything but DNA in dna.pdb and topol.top using text editor.

      3. Box information must be specified in dna.pdb. If it is not specified, define it using gmx editconf.

         • My box information in dna.pdb:

CRYST1  115.000  115.000   68.000  90.00  90.00  60.00 P 1           1

1. 2. 3. • We need to have a pair of DNA. We will duplicate the existing DNA.

           1. gmx editconf -f dna.pdb -o dna2.pdb -translate 4 0 0

           2. Add dna2.pdb to the end of dna.pdb using text editor.

           3. Edit topol.top to specify that we have TWO DNA molecules.

[ molecules ]

; Compound        #mols

DNA                 2

1. 2. 4. Solvate.

         • gmx solvate -cp dna.pdb -cs -o dna_water.pdb -p topol.top

           1. dna_water.pdb contains DNA and water molecules.

           2. topol.top: updated to specify the number of water molecules.

      5. Add ions.

         • Prepare TPR file for genion.

         • gmx grompp -f mini.mdp -c dna_water.pdb -maxwarn 40

         • gmx genion -neutral -conc 0.15 -s topol.tpr -o dna_water_ions.pdb -p topol.top

      6. Minimize

         • gmx grompp -f mini.mdp -c dna_water_ions.pdb  -maxwarn 40

         • gmx mdrun -nt 4 -gpu_id 0

           1. confout.gro  contains the final structure of the minimization.

           2. But, confout.gro may have broken DNA molecules due to PBC.

           3. We can make the molecules whole again by using gmx trjconv

              1. gmx trjconv -f confout.gro -o conf.pdb -pbc mol -ur compact

      7. Equilibration under NVT. Equilibration = 안정화, 예열, 본격적으로 production 전 제 자리 잡도록 도와줌. 

         • In nvt.mdp file,

; Temperature coupling  

tcoupl                   = v-rescale

; pressure coupling     

Pcoupl                   = no

1. 2. 7. • gmx grompp -c conf.pdb -f nvt.mdp -maxwarn 40

         • gmx mdrun -s topol.tpr -nt 8 -gpu_id 0 -deffnm nvt

      8. Equilibration under NpT. 

         • In npt.mdp file,

; Temperature coupling  

tcoupl                   = v-rescale

; pressure coupling     

Pcoupl                   = c-rescale

Pcoupltype               = semiisotropic

; Time constant (ps), compressibility (1/bar) and reference P (bar)

tau-p                    = 1.0

compressibility          = 4.5e-5 4.5e-5

ref-p                    = 1      1

1. 2. 8. • gmx grompp -c conf.pdb -f nvt.mdp -maxwarn 40

         • gmx mdrun -s topol.tpr -nt 8 -gpu_id 0 -deffnm npt

         • To see the box fluctuation in npt.xtc,

           2. 1. gmx traj -f npt.xtc -ob box.xvg

              2. 
                 

PMF (potential of mean forces) simulation using pull codes in Gromacs

• We will start pull simulations using npt.gro as a starting structure.

• We will measure mean forces at a range from inter-DNA distance d=20Å to 34Å with a spacing of 2 Å: 22, 24, 26, 28, 30, 32, 34Å (7 windows in total).

• Make a directory pull20 and cd to pull20, which is a window at DNA-DNA distance = 20Å.

  • We will reuse topol.top and npt.gro from the equilibration.

ln -s ../topol.top

ln -s ../dna.itp 

ln -s ../dna.hbonds.itp 

ln -s ../dna.pbc.itp 

ln -s ../npt.gro

• • Make groups for DNA1 and DNA2 using gmx make_ndx

    • gmx make_ndx -f npt.gro

ri 1-40

ri 41-80

name 9 DNA1

name 10 DNA2

q

• • • Group file will be written to index.ndx.

  • Copy npt.mdp to pull.mdp

    • cp ../npt.mdp pull.mdp

    • Edit pull.mdp using text editor to add pull codes.

; COM PULLING          

pull                     = yes

; Cylinder radius for dynamic reaction force groups (nm)

pull-nstxout             = 50

pull-nstfout             = 50

; Number of pull groups 

pull-ngroups             = 2

; Number of pull coordinates

pull-ncoords             = 1

; Group and coordinate parameters

pull-group1-name         = DNA1 ; must be consistent with NDX file.

pull-group2-name         = DNA2 ; must be consistent with NDX file.

pull-coord1-type         = umbrella

pull-coord1-geometry     = distance

pull-coord1-groups       = 1 2

pull-coord1-dim          = Y Y N    ; distance using deltaX and deltaY.

pull-coord1-init         = 2.0  ; spring length in nm

pull-coord1-k            = 1000 ; spring constant

• grompp

  • gmx grompp -f pull.mdp -maxwarn 40 -c npt.gro -n index.ndx

• gmx mdrun -nt 8 -gpu_id 0

  • Outputs:

    • pullx.xvg: spring length as a function of time.

    • pullf.xvg: spring force as a function of time. (f = -kx)

• 
  
• 
  

• Repeat the same pull simulation at d=22Å window.

  • Copy (or rsync) everything from pull20 to pull22.

  • Edit pull.mdp to modify spring length to 2.2 nm.

pull-coord1-init         = 2.2  ; spring length in nm

• • Then, gmx grompp & gmx mdrun

  • 
    

• Repeat the same thing for remaining windows.

• Plot mean forces as a function of spring length, d.

• Integrate the mean force curve using pythong, origin, excel... => ∆G(d) = PMF.

  • Otherwise, you can use gmx wham to get more accurate PMF.

• Compare ∆G curves before and after subtracting the entropic contribution from the Jacobian factor: kT log(r).

  • Note that the entropic contribution in water-water PMF was 2 kT log(r).

•