OpenMM and GROMACS preparation

Choose one engine route after parameterisation. The 06-series notebooks are alternatives (06A, 06B, 06C); pick one workflow, as listed in the notebook catalogue.

OpenMM: AMBER topology route

  1. Obtain matching P3HB_4.prmtop and P3HB_4.inpcrd from the GAFF2 step.

  2. Choose a fresh output folder.

  3. Start with a deliberately short run after the source and engine checks pass:

from pathlib import Path
from iphasimulator.simulation_openmm_amber_runner import run_openmm_with_amber_topology

base = Path("examples/output/md_tests/P3HB_4")
outputs = run_openmm_with_amber_topology(
    base / "gaff2/P3HB_4.prmtop",
    base / "gaff2/P3HB_4.inpcrd",
    base / "openmm/dry_polymer",
    minimization_max_iterations=200,
    nvt_steps=100, npt_steps=100, production_steps=100,
    report_interval=10, platform_name="CPU",
)

That call executes minimisation and dynamics; it is not just a file writer. The runner uses periodic handling when box vectors are present and records NPT as skipped for nonperiodic input. Review stage logs, the final structure and openmm_summary.log before increasing the run length. These short settings are for a smoke test, not equilibration or convergence evidence. The optional notebook 05A_quick_check_openmm.ipynb runs this check (Optional quick check of 05A parameters: 05A → 05A_quick_check_openmm); its run flag is off by default.

Solvated OpenMM with Amber force fields is notebook 06C (optional; enzyme + polymer in water). It builds the system in OPC water with tleap (ff19SB for the protein) and writes staged OpenMM run files with 06B’s stage names and lengths. 06B reproduces the project’s research simulations. 06C is a different force-field setup, so results are not directly comparable with 06B.

GROMACS with CHARMM-GUI inputs

For CGenFF/CHARMM36m systems prepared in CHARMM-GUI, use notebook 06B (enzyme + PHA, Solution Builder package).

GROMACS from GAFF2/AMBER files (conversion helpers)

The package can also convert GAFF2/AMBER files to GROMACS; the polymer-only benchmark systems were prepared this way, and notebook 06A runs these steps for one system.

  1. Use the converter/preparation helpers to convert AMBER parameters with ParmEd and create a self-contained dry run folder.

  2. Validate topology includes, atom counts and the box relative to nonbonded cutoffs before any grompp/minimisation run.

  3. Generate the explicit-water/ion preparation scripts from the dry inputs. Review solvent group names and force-field/water compatibility.

  4. Run the generated solvation script deliberately, then check coordinate and topology atom counts and solvent/ion molecule counts.

  5. Carry the prepared folder to HPC execution.

The standard folders are under examples/output/md_tests/P3HB_4/gromacs/, with dry_polymer/ and solvated_polymer/ branches. The preparation module also contains a CHARMM-GUI-style staged workflow; that is distinct from an automated enzyme-complex builder. Solvation scripts reset intermediate files in their target folder, so keep original inputs and use a dedicated workflow folder.

Notebook links: 06A–06B. API: iphasimulator.conversion_amber_to_gromacs.convert_amber_to_gromacs(), iphasimulator.simulation_gromacs_runner.prepare_gromacs_run_folder(), iphasimulator.simulation_gromacs_runner.validate_gromacs_coordinate_topology_counts().

Optional 06C: conditions and generated files

Use the 06C guide and 06B–06C comparison for the required docked complex, same-oligomer MOL2/FRCMOD/SDF files, matched salt/box/output conditions and checkpoint compatibility. 06C’s system.prmtop/system.inpcrd are a new solvated-complex pair from tleap, not the dry 05A files.