# Recommended way to thermostat a system where one covalently-bound component is fully restrained **URL:** \ **Category:** User discussions\ **Created:** [December 4, 2025, 2:13pm UTC](https://gromacs.bioexcel.eu/t/recommended-way-to-thermostat-a-system-where-one-covalently-bound-component-is-fully-restrained/13009 "2025-12-04T14:13:16Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![alis](https://avatars.discourse-cdn.com/v4/letter/a/ed655f/32.png) [@alis](https://gromacs.bioexcel.eu/u/alis)\ **Post date:** [December 4, 2025, 2:13pm UTC](https://gromacs.bioexcel.eu/t/recommended-way-to-thermostat-a-system-where-one-covalently-bound-component-is-fully-restrained/13009/1 "2025-12-04T14:13:16Z")
GROMACS version: 2024.2 GROMACS modification: No Hello everyone, I’m running an MD simulation of a protein covalently bound to a graphene sheet. The protein and the sheet were parameterized separately, and I bonded them using the intermolecular\_interactions directive at the end of the topol.top file: [intermolecular\_interactions] [distance\_restraints] 8002 9835 1 0 1 0.146 0.147 0.148 1.0 I want the graphene sheet to remain rigid during the simulation, so I generated a position restraint file and applied it to all graphene atoms. My question is about temperature coupling. Normally, I use two tc-grps (solute / solvent). But here the solute contains a fully restrained component (the sheet) covalently bound to an unrestrained component (the protein). If I do not assign the sheet to any tc-grps, GROMACS gives the fatal error: X atoms are not part of any T-coupling groups I want the sheet to stay immobile, so I’m unsure whether it is better to keep only two temperature-coupling groups (Protein+graphene together, and solvent), or define three groups (Protein / graphene / solvent), even though the graphene is fully restrained. I’m also concerned that including a completely restrained group together with the protein in the same tc-group may bias the temperature calculation. What is the recommended way to thermostat a system where one covalently-bound component is fully restrained and should remain rigid? Thanks in advance for your help! ---
**Author:** ![obZehn](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/obzehn/32/2629_2.png) [@obZehn](https://gromacs.bioexcel.eu/u/obZehn)\ **Post date:** [December 4, 2025, 5:10pm UTC](https://gromacs.bioexcel.eu/t/recommended-way-to-thermostat-a-system-where-one-covalently-bound-component-is-fully-restrained/13009/2 "2025-12-04T17:10:54Z")
I am unsure about how you should proceed in terms of coupling, but if I remember correctly if you use a thermostat then all the atoms of the systems must be part of a coupling group (and only once, no doubles). The silver lining is that if you check the [mdp options](https://manual.gromacs.org/current/user-guide/mdp-options.html#mdp-tau-t) you see that you can set the coupling constant `tau-t` to `tau-t = -1` to have no coupling in that group. So in case you decide to have it uncoupled then you can split the system atoms into solute/solvent/sheet and assign -1 to the sheet to have no thermostatting on it. ---
**Author:** ![alis](https://avatars.discourse-cdn.com/v4/letter/a/ed655f/32.png) [@alis](https://gromacs.bioexcel.eu/u/alis)\ **Post date:** [December 5, 2025, 3:02pm UTC](https://gromacs.bioexcel.eu/t/recommended-way-to-thermostat-a-system-where-one-covalently-bound-component-is-fully-restrained/13009/3 "2025-12-05T15:02:49Z")
Thank you! It seems to be working