ChimeraX docs icon
Tool: Minimize Structure
Minimize Structure energy-minimizes an atomic structure.
It is also implemented as the
minimize command.
The atoms are assigned types and associated force field parameters
as described for Add Charges, and
then the minimization is carried out using the OpenMM
LocalEnergyMinimizer with the
LangevinIntegrator.
See also:
Build Structure,
tug,
mouse modes
tug and
minimize
Minimize Structure automatically runs
dockprep
to perform several tasks to prepare the structure, including:
- deleting water molecules
- repairing truncated sidechains
- adding hydrogens
- assigning partial charges
To skip running dockprep
or to use specific options of that command, use
minimize instead
of the graphical interface.
Minimize Structure can be opened from the
Structure Editing section of the Tools menu and
manipulated like other panels (more...).
Only a single atomic model can be specified;
to combine multiple atomic models into one beforehand,
see combine.
- Minimize until convergence (initial default on)
– whether to minimize to convergence regardless of
how many steps it takes. If this option is turned off, a
Maximum # steps can be specified, meaning to halt minimization
after the specified number of steps has occurred or convergence is reached,
whichever is sooner.
Convergence is defined as when the root-mean-square of all force components
falls to 10 or lower.
- Update structure during minimization (initial default on)
– whether to update the structure periodically during the
minimization; otherwise, the structure will not be updated
until the minimization ends.
The current update interval is every 100 steps, at the same time
the energy is reported in the status line.
- Report energies to Log (initial default off)
– whether to report the energy in the
Log,
not just transiently in the status line
OK initiates the process and dismisses the dialog,
whereas Cancel simply dismisses the dialog. Help
opens this page in the Help Viewer.
The settings last used are remembered across sessions as
user preferences.
UCSF Resource for Biocomputing, Visualization, and Informatics /
October 2025