# GROMACS 2021.1 release notes¶

This version was released on March 8th, 2021. These release notes document the changes that have taken place in GROMACS since the previous 2021 version, to fix known issues. It also incorporates all fixes made in version 2020.6 and earlier, which you can find described in the Release notes.

## Fixes where mdrun could behave incorrectly¶

### Fix MiMiC with virtual sites¶

It is likely that MiMiC with virtual sites has not worked correctly because the call to construct the sites was placed after the call that uses the sites. Now it should work, but we have not tested that it does.

Issue 3866

### Fix mass perturbation to dH/dlambda¶

The contribution for perturbed mass was missing in dH/dlambda. Note that this contribution was not missing from the foreign energy differences used for the Bennett acceptance ratio method.

Issue 3943

### Running AWH with a convolved potential and an FEP dimension gives wrong results.¶

The output PMF is wrong when using awh-potential = convolved with a pull dimension combined with an FEP dimension. The FEP dimension always uses an umbrella potential and the combination does not work properly. This has been disabled in grompp.

Issue 3946

### Remove velocity from partially frozen atoms in md-vv¶

md-vv would add some velocity to the frozen dimensions of partially frozen atoms during constraining. This did not lead to wrong trajectories, as the frozen dimensions of the positions are kept fixed during propagation. The non-zero velocities were, however, reported in trajectories and final configurations. They might also have lead to slightly wrong kinetic energies, since the reported kinetic energy is calculated after the velocities are constrained. All effects are expected to be relatively small, since they did not accumulate, as the velocities were regularly reset to zero once per step.

Issue 3849

## Fixes for gmx tools¶

### Fix periodic boundary conditions in analysis framework tools¶

There was a bug in the trajectory analysis framework which caused molecules that were broken over PBC not to be made whole. This would usually lead to obviously incorrect outliers in analysis output.

Issue 3900

### Fix range checking bug in gmx covar¶

A check was inverted causing range checking to be applied wrong.

Issue 3902

### Fix various bugs in gmx xpm2ps¶

Numerous minor issues were introduced in refactoring since GROMACS 5.1, now fixed.

Issue 3881

## Fixes that affect portability¶

### Fixed compilation on Cygwin¶

A GROMACS header file was not including the necessary standard header. A problem with the M_PI math constant defined only by POSIX and not by C++ was also worked around.

Issue 3890

### Improve grompp checks of AWH settings when sampling an FEP dimension¶

Ensure that the AWH sampling interval is compatible with nstcalcenergy when sampling an FEP dimension using AWH. This avoids crashes in the first AWH sampling step (step > 0) if the settings were not correct.

Issue 3922

## Miscellaneous¶

• Updated GROMACS logos