A Lower Energy Isn’t the Whole Story: Checking Your Minimized Structure

Energy minimization is often the first point where a prepared molecular system gets a practical stability check. It can relax steric clashes and poor local geometry—but a falling energy plot alone does not tell you whether the run has met its stopping target. Checking both the potential energy and the maximum force helps you decide whether it is reasonable to continue to equilibration, or whether the minimization needs another look.

This is a common source of uncertainty: a run finishes, the energy has decreased, and the structure looks plausible. Is it ready for the next step? Use the checks below to make that decision from the actual output rather than from the plot alone.

1. Confirm that the energy trend makes sense

After the job finishes, inspect the potential-energy plot in the Plots section of the Minimize tab. A steady downward trend toward convergence is a reassuring sign that the minimization is progressing. In the example below, the plot shows a steady convergence.

However, a smooth-looking curve is not by itself a pass/fail test. The curve describes how potential energy evolves; the stopping target is expressed in terms of maximum force. Read the output as well as the plot.

2. Check the two values that matter

• Potential energy, Epot: it should be negative. For a simple protein in water, the documentation gives an approximate order of magnitude of 10⁵–10⁶, depending on system size and the number of water molecules. Treat that as context, not a universal cutoff: the value depends on the system.
• Maximum force, Fmax: compare this with the energy-minimization tolerance, emtol, set in the advanced parameters. For example, emtol = 1000.0 sets a target Fmax no greater than 1000 kJ mol^-1 nm^-1.

Look for these values in the Output window. They are reported during the run and at the end of the output. This makes it possible to distinguish a run that reached its force target from one that simply used all of its allowed minimization steps.

3. If the run stops before meeting the target

A minimization can reach the maximum number of steps and produce a reasonable-looking Epot while still ending with Fmax > emtol. In that situation, don’t treat the energy value alone as proof that the system is ready for equilibration. The system may not be stable enough for further equilibration and simulation.

Instead, use the output to identify the issue and consider the documented next options: increase the number of minimization steps, run minimization again using the previous minimization result as input, decrease the minimization step size, or adjust minimization parameters such as the integrator. These are alternatives to evaluate—not automatic fixes. After changing the setup, check the output and force target again.

A practical before-you-continue checklist

1. Confirm that the potential-energy plot shows a sensible trend.
2. Check that Epot is negative, interpreting its scale in the context of your system.
3. Compare the final Fmax with the configured emtol.
4. If the target was not reached, investigate before proceeding to NVT equilibration.

For a single project, the results may be offered for import when the job finishes. You can import the whole trajectory, only the last frame, or a selected range, and choose the periodic-boundary treatment and centering. If you cancel the import, the results are not deleted and the plot is still generated. Results are also available in the configured results folder; energy-minimization folders use the suffix _em. For batch jobs, plots are generated for each subproject, so check the individual results rather than assuming every system behaved identically.

Once the potential energy and maximum force are acceptable for your system, continue with Step 3: NVT Equilibration. For the full workflow and interface details, see the GROMACS Wizard energy-minimization documentation.

Note: SAMSON and all SAMSON Extensions are free for non-commercial use. Get SAMSON.