Run PyMACS

A first-run path for new users.

The recommended first experience is the curated protein-ligand walkthrough. It shows the full rhythm of PyMACS without asking a new user to design a new simulation from scratch.

Before running

Know what needs to exist first.

PyMACS automates a difficult workflow, but the scientific inputs still matter. Start from a curated example, then change one thing at a time when moving to a new system.

GROMACS available on the machine or cluster where the simulation will run The PyMACS repository files, including MDP templates and force-field folders The cgenff environment for setup and ligand conversion The mdanalysis environment for simulation control, analysis, plotting, and reports Ligand CGenFF files when running ligand or cofactor systems

First run

Follow the workflow in order.

1

Create a clean run folder

Keep each system in its own directory so generated topologies, trajectories, CSV tables, plots, and logs stay reproducible.

mkdir -p RUNS/Example_CPD32 && cd RUNS/Example_CPD32
2

Prepare the system

Run Step 1 in the cgenff environment. This cleans the structure, builds topologies, places solvent and ions, and writes the energy-minimization input.

conda activate cgenff
python 1_AutomateGromacs.py --pdb CPD32_9G94.pdb
3

Run equilibration and production

Run Step 2 in the analysis/simulation environment. This executes minimization, NVT, NPT, and production MD.

conda activate mdanalysis
python 2_AutomateGromacs.py
4

Analyze the trajectory

Step 3A recenters the trajectory, calculates stability and flexibility metrics, extracts binding-pocket contacts, and prepares network-ready tables.

python 3A_AutomateGromacs.py
5

Render networks and reports

Step 3B creates ligand-residue network views when the required contact tables exist. Step 4 assembles the final figurebook.

python 3B_NETWORX.py
python 4PDF4MD.py

Before production MD

The three stabilization stages.

Interpretation checks

Do these before trusting the story.

Read mdrun.log after every stage; it records commands and setup decisions.
Inspect em.gro, nvt.gro, and npt.gro before trusting production output.
Check RMSD and Rg before interpreting contacts.
Treat contact maps as evidence of proximity and persistence, not direct binding free energy.
Review CGenFF penalty scores for ligands before publication-grade conclusions.