# COM motion removal group

**URL:** <https://gromacs.bioexcel.eu/t/com-motion-removal-group/3404>\
**Category:** User discussions\
**Created:** [January 5, 2022, 12:06pm UTC](https://gromacs.bioexcel.eu/t/com-motion-removal-group/3404 "2022-01-05T12:06:00Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![rainbow](https://avatars.discourse-cdn.com/v4/letter/r/7993a0/32.png) [@rainbow](https://gromacs.bioexcel.eu/u/rainbow)\
**Post date:** [January 5, 2022, 12:06pm UTC](https://gromacs.bioexcel.eu/t/com-motion-removal-group/3404/1 "2022-01-05T12:06:00Z")

</div>

GROMACS version: 2021.4  
GROMACS modification: No

Dear gmx users,

# Q1

Referring to the answer of justin tutorial or gmx forum,

When protein in water simulation,  
comm-grps = system  
tc-grps = protein non-protein

The manual says to make tc-grps and comm-grps the same.  
comm-grps = protein non-protein  
tc-grps = protein non-protein

So, I’m confused which group I should to use.

# Q2

In protein in water simulation,  
When there are a number of free ligands to simulate their binding process to the active site,  
Does these free ligands belong to “protein” or “non-protein” for tc-grps?

Also, does comm-grps set to system?

Kind Regards,

---

<div class="post-metadata">

**Author:** ![jalemkul](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/jalemkul/32/18_2.png) [@jalemkul](https://gromacs.bioexcel.eu/u/jalemkul)\
**Post date:** [January 5, 2022, 2:38pm UTC](https://gromacs.bioexcel.eu/t/com-motion-removal-group/3404/2 "2022-01-05T14:38:14Z")

</div>

> [@rainbow](#):
>
> The manual says to make tc-grps and comm-grps the same.

Can you point to where the manual says this? I do not see any such statement, nor do I ever recall this.

Setting `tc-grps = protein non-protein` in a simple aqueous system can lead to instability if the COM motion corrections collide with one another. I’ve seen that happen, and such a setting is inappropriate.

> [@rainbow](#):
>
> When there are a number of free ligands to simulate their binding process to the active site,  
> Does these free ligands belong to “protein” or “non-protein” for tc-grps?

By default, they will be in the Non-Protein group. There is no straightforward answer for how to apply temperature coupling in this case, as the ligands may come and go from binding sites, but I doubt there is any substantial error by coupling them with the rest of the Non-Protein group.