In this example, we will be sampling COM distances from 0.5 - 5.0 nm along the z-axis using roughly 0.2-nm spacing. The following example commands may or may not be literally correct (the frame numbers may differ), but will serve as an example as to how to run grompp on separate coordinate files to generate all 23 inputs (note as well that 23 is the amount of windows required to obtain 0.2-nm spacing over roughly 4.5 nm; in our original work, 31 asymmetric windows were used).

After having identified the initial configurations of the sampling windows, we can now prepare the umbrella sampling simulations. We will need to generate a number of input files in order to conduct each of the necessary simulations. For example, since we have identified 23 configurations along the reaction coordinate, that means we will need 23 different input files for 23 independent simulations. Start by running a brief NPT equilibration in each window using this .mdp file.

First grompp (input conf*.gro file names may differ based on your individual SMD simulation outcome):

gmx grompp -f npt_umbrella.mdp -c conf6.gro -p topol.top -r conf6.gro -n index.ndx -o npt0.tpr
...
gmx grompp -f npt_umbrella.mdp -c conf449.gro -p topol.top -r conf449.gro -n index.ndx -o npt22.tpr

Then mdrun:

gmx mdrun -deffnm npt0
...
gmx mdrun -deffnm npt22

To start umbrella sampling, you will simply have to call grompp to process this .mdp file for each of the now-equilibrated configurations. Many of the pulling parameters are the same as in the SMD procedure, with the notable exception of pull_coord1_rate, which has now been set to zero. We don't want to move the configuration along the reaction coordinate; instead we want to restrain it within a defined window of configurational space. Setting pull_coord1_start = yes means that the initial COM distance is the reference distance, and we do not have to define a reference (pull_coord1_init) separately for each configuration.

gmx grompp -f md_umbrella.mdp -c npt0.gro -t npt0.cpt -p topol.top -r npt0.gro -n index.ndx -o umbrella0.tpr
...
gmx grompp -f md_umbrella.mdp -c npt22.gro -t npt22.cpt -p topol.top -r npt22.gro -n index.ndx -o umbrella22.tpr

Now, each input file should be passed to mdrun for the actual data collection simulation. Once all of the simulations are complete, you can proceed to data analysis.

gmx mdrun -deffnm umbrella0
...
gmx mdrun -deffnm umbrella22

Mike Harms has contributed a Python script that automates this process, extracting coordinate files and setting up the grompp and mdrun commands to streamline this process. You can download his script and associated documentation here. Please contact Mike if you have any feedback regarding this script or its use. Thanks for the contribution, Mike!