normal Polarizable water : LINCS warning

  • nina
  • nina's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
8 years 1 week ago #444 by nina
nina created the topic: Polarizable water : LINCS warning
Hello all,

I am testing the polarizable water with a protein that has a long flexible tail, very charged, and I want to see how it behaves if polarizable water particles are used. I have previously ran a test with standard water molecules with MARTINI, and the simulation went until the end with no problems.
But now the test with Pol Waters fails with erros in angle and bond rotation:

Step 5660140, time 169804 (ps) LINCS WARNING
relative constraint deviation after LINCS:
max 191.010162 (between atoms 1816 and 1818) rms 8.007496
bonds that rotated more than 90 degrees:

that affects only the pol water molecules.
I used the standard input mdp found in the site, do you have an idea where this problem comes from?

Thank you very much

Please Log in or Create an account to join the conversation.

More
8 years 1 week ago #447 by durbignon
durbignon replied the topic: Polarizable water : LINCS warning
Could you check if the constraints in the itp are fine - did you perhaps change them to bonded potentials for the minimisation? What is the angle potential you are using - is it the standard in the force-field?

Please Log in or Create an account to join the conversation.

More
7 years 6 months ago #682 by joshua_roc
joshua_roc replied the topic: Polarizable water : LINCS warning
We've seen the same error. We have made sure that we are using the correct terms (we switched to bonded potentials for minimization, but them we switched back for MD).

Have you ever come across this and if so, how did you fix it?

Step 5, time 0.005 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.255362, max 20.289154 (between atoms 10063 and 10065)
bonds that rotated more than 90 degrees:
atom 1 atom 2 angle previous, current, constraint length
10063 10064 90.0 0.1418 2.9017 0.1400
Wrote pdb files with previous and current coordinates

Step 6, time 0.006 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 71425465739.062271, max 4920597544960.000000 (between atoms 13237 and 13238)
bonds that rotated more than 90 degrees:
atom 1 atom 2 angle previous, current, constraint length
11491 11493 90.0 0.1400 270108016640.0000 0.1400
10339 10341 90.1 0.1400 3.2049 0.1400
10624 10625 90.0 0.1400 237773963264.0000 0.1400
10624 10626 90.0 0.1400 263136493568.0000 0.1400
12679 12680 90.0 0.1400 6680755.0000 0.1400
12679 12681 90.0 0.1400 4857049.5000 0.1400
13771 13773 93.1 0.1400 165244862464.0000 0.1400
Wrote pdb files with previous and current coordinates

Please Log in or Create an account to join the conversation.

More
7 years 6 months ago #684 by xavier
xavier replied the topic: Polarizable water : LINCS warning
It is confusing here! I am not sure there is a difference between the following lines and the one reported first!

Anyways, as Durbignon suggest you should look at the definition of the constrains.

Could you also indicate if this is a systematic error, I mean does it happen often? What is the length and sequence of the flexible part? successions of Alanines and Glycines could be an issue.

XAvier

joshua_roc wrote: We've seen the same error. We have made sure that we are using the correct terms (we switched to bonded potentials for minimization, but them we switched back for MD).

Have you ever come across this and if so, how did you fix it?

Step 5, time 0.005 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.255362, max 20.289154 (between atoms 10063 and 10065)
bonds that rotated more than 90 degrees:
atom 1 atom 2 angle previous, current, constraint length
10063 10064 90.0 0.1418 2.9017 0.1400
Wrote pdb files with previous and current coordinates

Step 6, time 0.006 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 71425465739.062271, max 4920597544960.000000 (between atoms 13237 and 13238)
bonds that rotated more than 90 degrees:
atom 1 atom 2 angle previous, current, constraint length
11491 11493 90.0 0.1400 270108016640.0000 0.1400
10339 10341 90.1 0.1400 3.2049 0.1400
10624 10625 90.0 0.1400 237773963264.0000 0.1400
10624 10626 90.0 0.1400 263136493568.0000 0.1400
12679 12680 90.0 0.1400 6680755.0000 0.1400
12679 12681 90.0 0.1400 4857049.5000 0.1400
13771 13773 93.1 0.1400 165244862464.0000 0.1400
Wrote pdb files with previous and current coordinates

Please Log in or Create an account to join the conversation.

More
7 years 6 months ago #686 by joshua_roc
joshua_roc replied the topic: Polarizable water : LINCS warning
The problem is occurring with the polarizable water. I have no glycines or alanines in the systems I'm trying to simulate.

In the .itp file for polarizable water, I see that for minimization I have to replace constraints with stiff bonds:

;for minimization purposes replace constraints by stiff bonds:

I've done that and I switched back for production MD. But I still get lots of these LINCS warnings. Do I need to minimize further or am I missing something else?

Joshua

xavier wrote: It is confusing here! I am not sure there is a difference between the following lines and the one reported first!

Anyways, as Durbignon suggest you should look at the definition of the constrains.

Could you also indicate if this is a systematic error, I mean does it happen often? What is the length and sequence of the flexible part? successions of Alanines and Glycines could be an issue.

XAvier

Please Log in or Create an account to join the conversation.

More
7 years 6 months ago #687 by xavier
xavier replied the topic: Polarizable water : LINCS warning
This is indeed a bit strange. A longer minimization could help!

However, other questions:
1- which value for the integration time step do you use? 0.20 fs should be fine.
2- any particularity of your sequence that might give troubles? Lots of charges ...
3- have you tried to use the peptide sequence giving troubles alone in solution?
4- how long in the segment?
5- Is it defined as coil in the topology? The small force constant used to describe the backbone-backbone bond might give troubles.

XAvier.

joshua_roc wrote: The problem is occurring with the polarizable water. I have no glycines or alanines in the systems I'm trying to simulate.

In the .itp file for polarizable water, I see that for minimization I have to replace constraints with stiff bonds:

;for minimization purposes replace constraints by stiff bonds:

I've done that and I switched back for production MD. But I still get lots of these LINCS warnings. Do I need to minimize further or am I missing something else?

Joshua

xavier wrote: It is confusing here! I am not sure there is a difference between the following lines and the one reported first!

Anyways, as Durbignon suggest you should look at the definition of the constrains.

Could you also indicate if this is a systematic error, I mean does it happen often? What is the length and sequence of the flexible part? successions of Alanines and Glycines could be an issue.

XAvier

Please Log in or Create an account to join the conversation.

More
7 years 6 months ago #688 by joshua_roc
joshua_roc replied the topic: Polarizable water : LINCS warning
1) Both a 20 fs step and a 1 fs step give this warning. (I find that running a few picoseconds using a 1fs step can shake out issues that minimization can't fix)

2) My system is a 2:1 POPE:POPG bilayer with water and neutralizing salt. The bilayer has a lot of charge (every POPG is -1 charge). I don't have to have any peptides in the system to get these warnings.

The only LINCS constraints in the system are the water.

Thanks so much for your responses!

Please Log in or Create an account to join the conversation.

More
7 years 6 months ago #689 by xavier
xavier replied the topic: Polarizable water : LINCS warning
Ok, problems with a 1 fs time step indicate that you have a more obvious problem. This could be a relaxation problem and likely fixed by a small simulation with a 1fs time step.

Did the problem persist after the small run with 1 fs time step?

joshua_roc wrote: 1) Both a 20 fs step and a 1 fs step give this warning. (I find that running a few picoseconds using a 1fs step can shake out issues that minimization can't fix)

2) My system is a 2:1 POPE:POPG bilayer with water and neutralizing salt. The bilayer has a lot of charge (every POPG is -1 charge). I don't have to have any peptides in the system to get these warnings.

The only LINCS constraints in the system are the water.

Thanks so much for your responses!

Please Log in or Create an account to join the conversation.

More
1 month 3 weeks ago #7853 by fgrunewald
fgrunewald replied the topic: Polarizable water : LINCS warning
Hey,

Which time-step are you using?

Do you use the particle definitions of the EO bead for polarizable water? Those can be found (not very intuitively) here: www.cgmartini.nl/images/parameters/polym...ini_v2.2refP_PEO.itp .

Otherwise, how large is your system and what is the maximum force after minimization?

Best,
Fabian

Please Log in or Create an account to join the conversation.

More
1 month 3 weeks ago #7854 by peterkroon
peterkroon replied the topic: Polarizable water : LINCS warning

Please Log in or Create an account to join the conversation.

Time to create page: 0.098 seconds