Martini AWH: Drug–polymer complex does not permeate membrane with position restraints; complex break

User discussions
1

GROMACS version:
GROMACS modification: Yes/No

System description

I am performing Martini CG AWH simulations of a drug–polymer complex (BDQ–polymer) interacting with a POPC membrane.

  • Membrane: POPC bilayer

  • Model: Martini CG

  • Collective variable: distance (Z) between membrane COM and complex COM

  • Simulation length:

    • 500 ns (no position restraints)

    • 1.5 µs (with position restraints)

  • NPT equilibration: 10 ns

Problem description

I am encountering two mutually exclusive issues:

  1. With position restraints (on either drug or polymer):

    • The complex does not permeate the membrane

    • It remains stuck outside or at the interface even after 1.5 µs

  2. Without position restraints:

    • The complex begins to permeate

    • However, the drug–polymer complex breaks apart during membrane crossing

What I suspect

  • Applying position restraints may be over-constraining internal degrees of freedom, preventing the conformational changes required for permeation

  • Without restraints, the nonbonded interactions between drug and polymer may be too weak, leading to dissociation in the hydrophobic membrane core

AWH / pull setup

  • Reaction coordinate: Z-direction (membrane normal)

  • Geometry: direction

  • Pull vector: (0, 0, -1)

  • Sampling range: -4.5 to 4.5 nm

MDP parameters (production)

integrator              = md
dt                      = 0.005
nsteps                  = 50000000

cutoff-scheme           = Verlet
rlist                   = 1.1

coulombtype             = PME
epsilon_r               = 15
rcoulomb                = 1.1

vdw-type                = cutoff
rvdw                    = 1.1

tcoupl                  = v-rescale
tc-grps                 = System
ref_t                   = 310

pcoupl                  = parrinello-rahman 
pcoupltype              = semiisotropic
tau_p                   = 12.0
ref_p                   = 1.0 1.0

constraints             = none

pull                        = yes
pull-ncoords                = 1
pull-ngroups                = 2

pull-group1-name            = MEMB
pull-group2-name            = COMPLEX

pull-coord1-type            = external-potential
pull-coord1-potential-provider = awh
pull-coord1-geometry        = direction
pull-coord1-dim             = N N Y
pull-coord1-vec             = 0 0 -1

awh                         = yes
awh-nbias                   = 1
awh1-dim1-start             = -4.5
awh1-dim1-end               = 4.5
awh1-dim1-force-constant    = 500
awh1-dim1-diffusion         = 1e-4

Questions

  1. Is it expected that position restraints prevent membrane permeation in AWH simulations of complexes?

  2. What is the recommended way to maintain complex integrity during permeation?

    • Should I use distance restraints between drug and polymer instead of position restraints?

  3. Could this indicate an issue with Martini parametrization or interaction strength between drug and polymer?

  4. Is my AWH force constant (500 kJ/mol/nm²) too strong for this system?

Additional notes

  • The system behaves differently depending on restraints, suggesting a balance issue between:

    • internal stability of the complex

    • flexibility required for membrane insertion

Any guidance on best practices for complex permeation with Martini + AWH would be highly appreciated.