Re: FATAL ERROR: Duplicate bond when going from nvt to npt

From: Marlon Sidore (marlon.sidore_at_gmail.com)
Date: Mon Oct 09 2017 - 04:04:17 CDT

Sorry for the late answer.

These atoms are part of the CG protein. 526 is a backbone PHE bead,
527-528-529 are the PHE cycle and 530 is the backbone PRO bead of the next
amino acid.

I don't see any duplicate bond in the .psf
521 523 523 524 524 525 525 526
526 527 526 530 527 528 527 529
528 529 530 531 530 532 532 533
532 534 534 535 534 536 536 537

There is an additional elastic network on the protein and there is only one
bond implicating these atoms
bond 527 530 0.59751434 5.4656

>From nvt to npt, there is the additional pressure control
(langevinPiston on
langevinPistonTarget 1.01325 ;# in bar -> 1 atm
langevinPistonPeriod 2000
langevinPistonDecay 1000
langevinPistonTemp $temp)
and the removing of the position restraints on the backbone beads
(fixedAtoms off) - nothing else is different.

I suspect it's from removing the position restraints (namd probably ignores
bonds that involve fixed atoms ?). But I don't see -where- there are
duplicate bonds ...

Also, I seem to have messed up with the namd list and should have "answer
to all" earlier, I apologize. Adding namd-l to the mail.

Marlon Sidore

PhD Student
Laboratoire d'Ingénierie des Systèmes Macromoléculaire (LISM)
CNRS - UMR7255
31, Chemin Joseph Aiguier
13402 cedex 20 Marseille
France

2016-12-12 18:25 GMT+01:00 Radak, Brian K <bradak_at_anl.gov>:

> What are atoms 526 and 530? Are they part of a lipid?
>
> Brian Radak
> Postdoctoral Appointee
> Leadership Computing Facility
> Argonne National Laboratory
>
> 9700 South Cass Avenue, Bldg. 240
> Argonne, IL 60439-4854
> (630) 252-8643
> brian.radak_at_anl.gov
> ------------------------------
> *From:* Marlon Sidore [marlon.sidore_at_gmail.com]
> *Sent:* Monday, December 12, 2016 11:19 AM
> *To:* Radak, Brian K
> *Subject:* Re: namd-l: FATAL ERROR: Duplicate bond when going from nvt to
> npt
>
> I have managed to go on until the production run ... while changing
> absolutely nothing to the parameters (did 2 npt, one with position
> restraints and one without). The production run now gives the same error. I
> am not sure about which settings are relevant. I also have checked for
> duplicate atoms.
>
> Info: Reading from binary file npt2.restart.coor
> Info:
> Info: Entering startup at 1.9668 s, 456.715 MB of memory in use
> Info: Startup phase 0 tppl 0.000170946 s, 456.715 MB of memory in use
> FATAL ERROR: Duplicate bond rom atom 526 tp atom 530
> FATAL ERROR: Duplicate bond rom atom 526 tp atom 530
> FATAL ERROR: Duplicate bond rom atom 526 tp atom 530
> FATAL ERROR: Duplicate bond rom atom 526 tp atom 530
> FATAL ERROR: Duplicate bond rom atom 526 tp atom 530
> ... And so on for a few more dozen lines.
>
> Marlon Sidore
>
>
> PhD Student
> Laboratoire d'Ingénierie des Systèmes Macromoléculaire (LISM)
> CNRS - UMR7255
> 31, Chemin Joseph Aiguier
> 13402 cedex 20 Marseille
> France
>
>
> 2016-12-12 16:33 GMT+01:00 Radak, Brian K <bradak_at_anl.gov>:
>
>> Can you post the exact error and the relevant settings?
>>
>> Brian Radak
>> Postdoctoral Appointee
>> Leadership Computing Facility
>> Argonne National Laboratory
>>
>> 9700 South Cass Avenue, Bldg. 240
>> Argonne, IL 60439-4854
>> (630) 252-8643
>> brian.radak_at_anl.gov
>> ------------------------------
>> *From:* owner-namd-l_at_ks.uiuc.edu [owner-namd-l_at_ks.uiuc.edu] on behalf of
>> Marlon Sidore [marlon.sidore_at_gmail.com]
>> *Sent:* Monday, December 12, 2016 4:02 AM
>> *To:* NAMD list
>> *Subject:* namd-l: FATAL ERROR: Duplicate bond when going from nvt to npt
>>
>> Hello,
>>
>> I am currently doing a simulation of a membrane protein. The NVT happened
>> smoothly, with position restraints on the backbone of the protein.
>> The error appears when activating the pressure control (while keeping the
>> position restraints). I don't understand this error though, since it ran
>> fine in NVT - and starting NPT from the result of the NVT.
>>
>> Why, if there is indeed a duplicate bond in my topology (though I
>> couldn't find it), did it run smoothly in NVT ? What did I miss ?
>>
>> Thanks
>>
>> Marlon Sidore
>>
>>
>> PhD Student
>> Laboratoire d'Ingénierie des Systèmes Macromoléculaire (LISM)
>> CNRS - UMR7255
>> 31, Chemin Joseph Aiguier
>> 13402 cedex 20 Marseille
>> France
>>
>>
>

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2018 - 23:20:38 CST