Re: Re: Re: QM atom cannot be fixed in space

From: Alex Balaeff (abalaeff_at_polarisqb.com)
Date: Tue Sep 01 2020 - 23:47:18 CDT

Zhihong:

1) Make sure you create that directory, /dev/shm/NAMD_Example1 prior to
running NAMD, that the directory is writable, there is enough space in
/dev/shm, etc.

2) Make sure the Orca executable is indeed in the path you indicated, is
executable in Linux, etc.

3) If it's still crashing after 1) and 2), collect the input files that
NAMD created in the /dev/shm/NAMD_Example1/0 directory and run a standalone
Orca calculation with those files. Perhaps, something wrong with the QM
system is causing ORCA to crash.

Best,

Alexander.

On Mon, Aug 31, 2020 at 9:21 AM 辛志宏 <xzhfood_at_njau.edu.cn> wrote:

> Dear Alexander,
> Big thanks for your kindly help.
> I double check the QM region and reselect the QM atom,and setup constant
> to 5 kcal/(mol*A^2) for backbone atom:
>
> set all [atomselect top all]
> set sel [atomselect top "protein and backbone"]
> $all set beta 0.0
> $sel set beta 1.0
> $all writepdb YZZ-restraint.pdb
>
> in the namd config file:
>
> constraints on
> consexp 2
> consref ./YZZ-restraint.pdb
> conskfile ./YZZ-restraint.pdb
> conskcol B
> constraintScaling 5.0
>
> I also save QM region as a separate file, now, it seems everything is
> fine, but a new issue occured,
>
> [file orca_main/maininp2.cpp, line 414]: Error: number expected in
> COORDS/FirstCoordinate ATOM-NO 114
>
> in the *.log file, the error information is as follows:
>
> TCL: Minimizing for 100 steps
>
> Info: List of ranks running QM simulations: 0.
>
> FATAL ERROR: Error running command for QM forces calculation.
>
> I searched the namd mailing list, but did not find the solution, I checked
> the qmBaseDir "/dev/shm/NAMD_Example1" to try find more information, there
> is nothing.
>
> what should I do to fix it? Thank you for your help.
>
> I attached the namd config file and log file here for detail information.
>
> Best Wishes,
> Zhihong Xin,
>
>
> From: Alex Balaeff <abalaeff_at_polarisqb.com>
> Date: 2020-08-31 13:41:26
> To: "辛志宏" <xzhfood_at_njau.edu.cn>
> Cc: namd-l_at_ks.uiuc.edu
> Subject: Re: Re: namd-l: QM atom cannot be fixed in space
>
> Zhihong:
>
> What error did you get? Please include the log file with the error
> description and your namd script.
>
> Just looking at the file you've sent, I have the following comments:
>
> 1) Your beta column is set to 1 which indicates to NAMD that you want to
> restrain those atoms to their indicated positions with the force constant
> of 1 kcal/(mol*A^2). I suggest setting the constant to 5 or 10
> kcal/(mol*A^2)
>
> 2) Don't restrain hydrogens. Restraining the heavy atoms is quite enough.
> Unless the Hs in your simulation move wildly and you do need to restain
> them as well.
>
> 3) Do use separate PDB files to flag the restrained atoms and QM atoms. In
> the QM file, you need both beta and occupancy columns, since your QM
> subsystem seems to be covalently bound to the MM subsystem.
>
> 4) Your QM system selection seems to include water molecules ## 75, 76,
> etc.; is that what you wanted? If not, edit your VMD selection:
>
> set QM [atomselect top "(protein and resid 75 76 77 144 193 238 268) or
> (resname DMP) or (resname TIP3 and resid 4231)"]
>
> (please double-check whether this selection does include every residue you
> wanted into the QM region and correct as needed)
>
>
> Best,
>
> Alexander.
>
>
> On Sun, Aug 30, 2020 at 10:32 AM 辛志宏 <xzhfood_at_njau.edu.cn> wrote:
>
>>
>> Thank you Alexander,
>> I restrain the QM atom as follows,but a new error occured regarding to
>> QM atom assignment, how to fixed it ? I attached the my QM file here.
>> constraints on
>> consexp 2
>> consref ./YZZ-QMMM_fixed.pdb
>> conskfile ./YZZ-QMMM_fixed.pdb
>> conskcol B
>>
>> Thank you again.
>> Zhihong Xin
>>
>>
>>
>>
>> From: Alex Balaeff <abalaeff_at_polarisqb.com>
>> Date: 2020-08-29 22:29:48
>> To: namd-l_at_ks.uiuc.edu,"辛志宏" <xzhfood_at_njau.edu.cn>
>> Subject: Re: namd-l: QM atom cannot be fixed in space
>>
>> Zhihong:
>>
>> My suggestion is to harmonically restrain the QM atom rather than to keep
>> it fixed. If you want to limit the movement of your QM molecule around the
>> simulation cell, the effect is going to be the same. See
>> https://www.ks.uiuc.edu/Research/namd/2.14/ug/node29.html , "Position
>> restraint"
>>
>> Also I suggest to keep the QM region PDB and the restraint PDB as two
>> separate files. Makes it easier to change things if needed (e.g., if you
>> need to gradually lift the harmonic restraint).
>>
>> Best,
>>
>> Alexander.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> On Sat, Aug 29, 2020 at 8:53 AM 辛志宏 <xzhfood_at_njau.edu.cn> wrote:
>>
>>> Dear ALL,
>>> An ERROR occured when I run QM/MM by NAMD, the error information and the
>>> QM region setup as follows:
>>>
>>> 。。。。。。。。。。。。。。。。。。。
>>>
>>> ERROR: QM atom cannot be fixed in space!
>>> FATAL ERROR: Error processing QM information.
>>>
>>>
>>>
>>> QM region setup:
>>>
>>> set allatoms [atomselect top all]
>>> $allatoms set beta 0
>>> $allatoms set occupancy 0
>>> set QM [atomselect top "resid 75 76 77 144 193 238 268 299 4231"]
>>> $QM set beta 1
>>> $QM set occupancy 1
>>> $allatoms writepdb YZZ-namd-QM.pdb
>>>
>>>
>>>
>>> Any suggestion will be much appreciate for fixing the issue, thank you
>>> in advance.
>>>
>>>
>>> Zhihong XIn
>>>
>>>
>>>
>>
>> --
>> -----
>> Dr. Alexander Balaeff
>> * Polaris Quantum Biotech www.PolarisQB.com <http://www.PolarisQB.com>*
>> (919)-270-5772
>>
>>
>>
>
> --
> -----
> Dr. Alexander Balaeff
> * Polaris Quantum Biotech www.PolarisQB.com <http://www.PolarisQB.com>*
> (919)-270-5772
>
>
>

-- 
 -----
  Dr. Alexander Balaeff
*  Polaris Quantum Biotech  www.PolarisQB.com <http://www.PolarisQB.com>*
  (919)-270-5772

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2020 - 23:17:14 CST