# Longer NPT equilibration

**URL:** <https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349>\
**Category:** User discussions\
**Created:** [June 10, 2024, 2:29am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349 "2024-06-10T02:29:37Z")\
**Posts on this page:** 13\
**Page:** 3

<div class="post-metadata">

**Author:** ![MagnusL](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/magnusl/32/2380_2.png) [@MagnusL](https://gromacs.bioexcel.eu/u/MagnusL)\
**Post date:** [July 2, 2024, 3:16pm UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/42 "2024-07-02T15:16:34Z")

</div>

I’m not sure exactly what you mean. If you have found that your equilibration simulation is long enough to be used for production, you can indeed use `gmx energy -b <time>` with a proper starting time and use those values for measurements. In principle, if your settings are the same in the NPT equilibration as in the production simulation, you can just run one long simulation and discard the beginning as equilibration, e.g., by using `gmx energy -b <time>`.

---

<div class="post-metadata">

**Author:** ![SP06](https://avatars.discourse-cdn.com/v4/letter/s/ecb155/32.png) [@SP06](https://gromacs.bioexcel.eu/u/SP06)\
**Post date:** [July 2, 2024, 3:20pm UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/43 "2024-07-02T15:20:52Z")

</div>

I meant to say, I want to calculate the density and pressure in the range 3-10 or 3-15 ns from the entire 50 ns run.

---

<div class="post-metadata">

**Author:** ![MagnusL](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/magnusl/32/2380_2.png) [@MagnusL](https://gromacs.bioexcel.eu/u/MagnusL)\
**Post date:** [July 2, 2024, 3:41pm UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/44 "2024-07-02T15:41:40Z")

</div>

Then, use `gmx energy -b <start time> -e <end time>`

---

<div class="post-metadata">

**Author:** ![SP06](https://avatars.discourse-cdn.com/v4/letter/s/ecb155/32.png) [@SP06](https://gromacs.bioexcel.eu/u/SP06)\
**Post date:** [July 3, 2024, 5:52am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/45 "2024-07-03T05:52:09Z")

</div>

The barostat used for NPT and production are different (while other settings are same), also pos.restraints relaxed in production. In NPT, Berendsen barostat was used (10 ns equilibration) while for production Parrinello-Rahman. But the system pressure got stabilized during the production run and was stable for each 100 ns run. Are the results still valid with different barostats? Or, do I need to prepare that system again?

---

<div class="post-metadata">

**Author:** ![MagnusL](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/magnusl/32/2380_2.png) [@MagnusL](https://gromacs.bioexcel.eu/u/MagnusL)\
**Post date:** [July 3, 2024, 9:14am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/46 "2024-07-03T09:14:44Z")

</div>

Yes, you can switch barostats between equilibration and production. Just start measuring your observables when you are certain that the system is stable (in equilibrium) by discarding the beginning of the production simulation.

---

<div class="post-metadata">

**Author:** ![SP06](https://avatars.discourse-cdn.com/v4/letter/s/ecb155/32.png) [@SP06](https://gromacs.bioexcel.eu/u/SP06)\
**Post date:** [July 4, 2024, 8:25am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/47 "2024-07-04T08:25:47Z")

</div>

Sir. I had confusion about position restraints. In a protein-solvent system, we are applying restraints to the protein during equilibration steps. It’s said that -DPOSRE in the nvt.mdp and npt.mdp restrains all atoms other than hydrogen. I wanted to know while we are equilibrating a box of cyclohexane, do we need to apply position restraints there too?

---

<div class="post-metadata">

**Author:** ![MagnusL](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/magnusl/32/2380_2.png) [@MagnusL](https://gromacs.bioexcel.eu/u/MagnusL)\
**Post date:** [July 4, 2024, 8:51am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/48 "2024-07-04T08:51:11Z")

</div>

In a protein-solvent system you usually use restraints (in early equilibration stages) to keep the protein reasonably close to its initial configuration, mainly so that any instabilities during the equilibration does not distort the protein too much. Restraints (possibly also of the solvent) may also be necessary to keep the system from crashing.

So, what you need to ask yourself is, are you particularly interested in the starting configuration of the cyclohexane box? If so, then you should apply restraints. Is the system crashing without restraints? If so, see if it helps using restraints.

There are few general rights and wrongs in MD simulations. In certain situations some things are good to do, whereas in others they are not. As long as you understand why you are using certain algorithms and/or applying certain limitations or restraints (in this case I don’t mean specifically coordinate restraints), you will probably be fine.

---

<div class="post-metadata">

**Author:** ![SP06](https://avatars.discourse-cdn.com/v4/letter/s/ecb155/32.png) [@SP06](https://gromacs.bioexcel.eu/u/SP06)\
**Post date:** [July 4, 2024, 9:49am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/49 "2024-07-04T09:49:09Z")

</div>

Sir, as I have said earlier I am preparing this cyclohexane box to prepare the cyclohexane layer for a biphase (water-cyclohexane) , just following the tutorial. So, what will you suggest about applying restraints here? As I am preparing biphasic system for the first time, I do not understand many things. So, can you please suggest me what should be done in this case.

---

<div class="post-metadata">

**Author:** ![MagnusL](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/magnusl/32/2380_2.png) [@MagnusL](https://gromacs.bioexcel.eu/u/MagnusL)\
**Post date:** [July 4, 2024, 10:02am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/50 "2024-07-04T10:02:02Z")

</div>

Then I would say that “Is the system crashing without restraints? If so, see if it helps using restraints.” applies.

And even if you are learning, always try to understand why things are done the way they are in tutorials etc.

---

<div class="post-metadata">

**Author:** ![SP06](https://avatars.discourse-cdn.com/v4/letter/s/ecb155/32.png) [@SP06](https://gromacs.bioexcel.eu/u/SP06)\
**Post date:** [July 4, 2024, 11:40am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/51 "2024-07-04T11:40:03Z")

</div>

Yes Sir, I understand.

---

<div class="post-metadata">

**Author:** ![SP06](https://avatars.discourse-cdn.com/v4/letter/s/ecb155/32.png) [@SP06](https://gromacs.bioexcel.eu/u/SP06)\
**Post date:** [July 10, 2024, 5:06pm UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/53 "2024-07-10T17:06:29Z")

</div>

Sir, can C-rescale be used for semi-isotropic systems too for NPT and production run, or P-R is better?

---

<div class="post-metadata">

**Author:** ![MagnusL](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/magnusl/32/2380_2.png) [@MagnusL](https://gromacs.bioexcel.eu/u/MagnusL)\
**Post date:** [July 12, 2024, 7:14am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/54 "2024-07-12T07:14:22Z")

</div>

I would recommend c-rescale in all cases except anisotropic pressure coupling, for which it does not work yet.

---

<div class="post-metadata">

**Author:** ![SP06](https://avatars.discourse-cdn.com/v4/letter/s/ecb155/32.png) [@SP06](https://gromacs.bioexcel.eu/u/SP06)\
**Post date:** [July 15, 2024, 10:07am UTC](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349/56 "2024-07-15T10:07:26Z")

</div>

Sir, after NPT equilibration of a cyclohexane box, I extended it in z-direction, placed protein, solvated with cyclohexane and water (as per tutorial). changed vdwradii of C to 0.35. Then when I went on for Energy minimization I got following error -  
Steepest Descents:  
Tolerance (Fmax) = 1.00000e+03  
Number of steps = 50000

WARNING: Listed nonbonded interaction between particles 29881 and 29890  
at distance 4.043 which is larger than the table limit 2.200 nm.

This is likely either a 1,4 interaction, or a listed interaction inside  
a smaller molecule you are decoupling during a free energy calculation.  
Since interactions at distances beyond the table cannot be computed,  
they are skipped until they are inside the table limit again. You will  
only see this message once, even if it occurs for several interactions.

IMPORTANT: This should not happen in a stable simulation, so there is  
probably something wrong with your system. Only change the table-extension  
distance in the mdp file if you are really sure that is the reason.

WARNING: Listed nonbonded interaction between particles 6805 and 6812  
at distance 4.018 which is larger than the table limit 2.200 nm.

This is likely either a 1,4 interaction, or a listed interaction inside  
a smaller molecule you are decoupling during a free energy calculation.  
Since interactions at distances beyond the table cannot be computed,  
they are skipped until they are inside the table limit again. You will  
only see this message once, even if it occurs for several interactions.

IMPORTANT: This should not happen in a stable simulation, so there is  
probably something wrong with your system. Only change the table-extension  
distance in the mdp file if you are really sure that is the reason.

WARNING: Listed nonbonded interaction between particles 26677 and 26686  
at distance 4.307 which is larger than the table limit 2.200 nm.

This is likely either a 1,4 interaction, or a listed interaction inside  
a smaller molecule you are decoupling during a free energy calculation.  
Since interactions at distances beyond the table cannot be computed,  
they are skipped until they are inside the table limit again. You will  
only see this message once, even if it occurs for several interactions.

IMPORTANT: This should not happen in a stable simulation, so there is  
probably something wrong with your system. Only change the table-extension  
distance in the mdp file if you are really sure that is the reason.

WARNING: Listed nonbonded interaction between particles 10441 and 10450  
at distance 4.307 which is larger than the table limit 2.200 nm.

This is likely either a 1,4 interaction, or a listed interaction inside  
a smaller molecule you are decoupling during a free energy calculation.  
Since interactions at distances beyond the table cannot be computed,  
they are skipped until they are inside the table limit again. You will  
only see this message once, even if it occurs for several interactions.

IMPORTANT: This should not happen in a stable simulation, so there is  
probably something wrong with your system. Only change the table-extension  
distance in the mdp file if you are really sure that is the reason.

**But with -ntmpi 1, energy minimization is done, but for 2-3 steps I got 1/more water molecules not settled. When I proceeded to NVT, again I got following error -**

Step 0, time 0 (ps) LINCS WARNING  
relative constraint deviation after LINCS:  
rms 0.000006, max 0.000320 (between atoms 38632 and 38634)  
bonds that rotated more than 30 degrees:  
atom 1 atom 2 angle previous, current, constraint length  
38632 38633 36.0 0.1111 0.1111 0.1111  
step 0  
Step 1, time 0.002 (ps) LINCS WARNING  
relative constraint deviation after LINCS:  
rms 0.001882, max 0.131746 (between atoms 38632 and 38633)  
bonds that rotated more than 30 degrees:  
atom 1 atom 2 angle previous, current, constraint length  
38632 38633 90.0 0.1111 0.1257 0.1111

Step 2, time 0.004 (ps) LINCS WARNING  
relative constraint deviation after LINCS:  
rms 0.020097, max 1.407050 (between atoms 38632 and 38633)  
bonds that rotated more than 30 degrees:  
atom 1 atom 2 angle previous, current, constraint length  
34003 34004 43.4 0.1111 0.1110 0.1111  
38632 38633 90.0 0.1257 0.2674 0.1111  
Wrote pdb files with previous and current coordinates

Step 3, time 0.006 (ps) LINCS WARNING  
relative constraint deviation after LINCS:  
rms 0.001757, max 0.120917 (between atoms 38632 and 38633)  
bonds that rotated more than 30 degrees:  
atom 1 atom 2 angle previous, current, constraint length  
34003 34004 36.5 0.1110 0.1110 0.1111  
38632 38633 90.0 0.2674 0.1245 0.1111

Step 4, time 0.008 (ps) LINCS WARNING  
relative constraint deviation after LINCS:  
rms 0.357965, max 18.061623 (between atoms 38632 and 38633)  
bonds that rotated more than 30 degrees:  
atom 1 atom 2 angle previous, current, constraint length  
38635 38636 90.0 0.1111 0.1307 0.1111  
38635 38637 90.0 0.1111 2.0396 0.1111  
34003 34004 90.0 0.1110 0.1538 0.1111  
39397 39399 90.0 0.1111 0.1829 0.1111  
38632 38633 90.0 0.1245 2.1177 0.1111  
38632 38634 90.0 0.1136 0.1211 0.1111  
Wrote pdb files with previous and current coordinates

Step 5, time 0.01 (ps) LINCS WARNING  
relative constraint deviation after LINCS:  
rms 54.655209, max 3485.760498 (between atoms 39397 and 39399)  
bonds that rotated more than 30 degrees:  
atom 1 atom 2 angle previous, current, constraint length  
39703 39704 90.0 0.1111 0.1709 0.1111  
38635 38636 90.0 0.1307 0.3210 0.1111  
38635 38637 62.7 2.0396 0.1042 0.1111  
34003 34004 90.0 0.1538 0.1344 0.1111  
39397 39398 90.0 0.1114 9.6527 0.1111  
39397 39399 90.0 0.1829 387.3791 0.1111  
38632 38633 90.0 2.1177 0.1328 0.1111  
38632 38634 90.0 0.1211 0.2848 0.1111  
39400 39401 90.0 0.1111 172.9535 0.1111  
39400 39402 90.0 0.1111 28.4335 0.1111  
Wrote pdb files with previous and current coordinates

Program: gmx mdrun, version 2021.4-Ubuntu-2021.4-2  
Source file: src/gromacs/ewald/pme\_redistribute.cpp (line 305)  
MPI rank: 19 (out of 20)

Fatal error:  
1 particles communicated to PME rank 1 are more than 2/3 times the cut-off out  
of the domain decomposition cell of their charge group in dimension x.  
This usually means that your system is not well equilibrated.

Kindly help me resolve this issue. I got cyclohexane topology and coordinate files generated by charmm\_GUI. Please note- I am not doing free energy calculation.  
Can you kindly help me solve this?

Also, I wanted to ask, for this biphasic system with peptide placed, should I consider 2 tc\_grps= protein non-protein or one tc\_grp= System?

[Previous page](https://gromacs.bioexcel.eu/t/longer-npt-equilibration/9349.md?page=2)