Error in csg_inverse step

56 views
Skip to first unread message

Souvik Chakraborty

unread,
Aug 14, 2019, 8:40:19 AM8/14/19
to votca
Dear group members,

While running csg_inverse to get non-bonded potential, I get the following error in inverse.log:

Fatal error:
1 of the 4500 bonded interactions could not be calculated because some atoms
involved moved further apart than the multi-body cut-off distance (1.5 nm) or
the two-body cut-off distance (1.5 nm), see option -rdd, for pairs and
tabulated bonds also see option -ddcheck


######################################################################

How to solve this error? Is there any way to customize the 'gmx mdrun' command inside the scripts?


Another issue: How to include exclusions in topol.top?
(I have created the file exclusions.txt, and included it in topol.top. But, csg_inverse cannot find it)


Thanks in advance!

-Souvik

Christoph Junghans

unread,
Aug 14, 2019, 1:28:58 PM8/14/19
to vo...@googlegroups.com
On Wed, Aug 14, 2019 at 6:40 AM Souvik Chakraborty <souv...@gmail.com> wrote:
>
> Dear group members,
>
> While running csg_inverse to get non-bonded potential, I get the following error in inverse.log:
>
> Fatal error:
> 1 of the 4500 bonded interactions could not be calculated because some atoms
> involved moved further apart than the multi-body cut-off distance (1.5 nm) or
> the two-body cut-off distance (1.5 nm), see option -rdd, for pairs and
> tabulated bonds also see option -ddcheck
>
>
> ######################################################################
>
> How to solve this error?
This usually means a bond got stretched too far, so something isn't
setup correctly.

> Is there any way to customize the 'gmx mdrun' command inside the scripts?
Yes. It is the cg.inverse.gromacs.mdrun.command option in the xml
file. (see sec. 10.5.83 of http://doc.votca.org/manual.pdf)

>
>
> Another issue: How to include exclusions in topol.top?
> (I have created the file exclusions.txt, and included it in topol.top. But, csg_inverse cannot find it)
Did you forget to exclusions.txt add to the <filelist>?

Christoph
>
>
> Thanks in advance!
>
> -Souvik
>
> --
> You received this message because you are subscribed to the Google Groups "votca" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to votca+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/ad305ca4-49c4-4017-9690-98f9bf8a8e6a%40googlegroups.com.



--
Christoph Junghans
Web: http://www.compphys.de

Souvik Chakraborty

unread,
Aug 15, 2019, 2:23:56 AM8/15/19
to votca
Hi Christoph,

Thanks for the response.

I want to add -rdd flag. Maybe a large -rdd can solve the error. I am not clear how to modify settings.xml to apply that.

Please shed some light!

Kind regards,

Souvik

Souvik Chakraborty

unread,
Aug 15, 2019, 8:36:41 AM8/15/19
to votca

Update: I could figure out how to include flags to 'gmx mdrun' in settings.xml. The error is still the same. I probably need to energy-minimize the conf.gro well before csg_inverse.

Regards,
Souvik

On Wednesday, 14 August 2019 20:40:19 UTC+8, Souvik Chakraborty wrote:

Christoph Junghans

unread,
Aug 15, 2019, 8:49:41 AM8/15/19
to vo...@googlegroups.com
On Thu, Aug 15, 2019 at 12:23 AM Souvik Chakraborty <souv...@gmail.com> wrote:
>
> Hi Christoph,
>
> Thanks for the response.
>
> I want to add -rdd flag. Maybe a large -rdd can solve the error. I am not clear how to modify settings.xml to apply that.
If you just want to add an option you can use the
cg.inverse.gromacs.mdrun.opts for that.

Christoph
>
> Please shed some light!
>
> Kind regards,
>
> Souvik
>
> On Wednesday, 14 August 2019 20:40:19 UTC+8, Souvik Chakraborty wrote:
>>
>> Dear group members,
>>
>> While running csg_inverse to get non-bonded potential, I get the following error in inverse.log:
>>
>> Fatal error:
>> 1 of the 4500 bonded interactions could not be calculated because some atoms
>> involved moved further apart than the multi-body cut-off distance (1.5 nm) or
>> the two-body cut-off distance (1.5 nm), see option -rdd, for pairs and
>> tabulated bonds also see option -ddcheck
>>
>>
>> ######################################################################
>>
>> How to solve this error? Is there any way to customize the 'gmx mdrun' command inside the scripts?
>>
>>
>> Another issue: How to include exclusions in topol.top?
>> (I have created the file exclusions.txt, and included it in topol.top. But, csg_inverse cannot find it)
>>
>>
>> Thanks in advance!
>>
>> -Souvik
>>
> --
> You received this message because you are subscribed to the Google Groups "votca" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to votca+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/d5829ca2-c161-4506-8a33-083fc7f96fd6%40googlegroups.com.

Christoph Junghans

unread,
Aug 15, 2019, 5:24:40 PM8/15/19
to vo...@googlegroups.com
On Thu, Aug 15, 2019 at 6:36 AM Souvik Chakraborty <souv...@gmail.com> wrote:
>
>
> Update: I could figure out how to include flags to 'gmx mdrun' in settings.xml. The error is still the same. I probably need to energy-minimize the conf.gro well before csg_inverse.

One can do an energy minimization before every step if needed, we call
it a pre-simulation, see here:
https://github.com/votca/csg-tutorials/blob/master/spce/simplex/simple/settings.xml#L57-L59
You just need to specify an additional mdp file (and/or topology file
if needed).

Christoph
>
> Regards,
> Souvik
>
> On Wednesday, 14 August 2019 20:40:19 UTC+8, Souvik Chakraborty wrote:
>>
>> Dear group members,
>>
>> While running csg_inverse to get non-bonded potential, I get the following error in inverse.log:
>>
>> Fatal error:
>> 1 of the 4500 bonded interactions could not be calculated because some atoms
>> involved moved further apart than the multi-body cut-off distance (1.5 nm) or
>> the two-body cut-off distance (1.5 nm), see option -rdd, for pairs and
>> tabulated bonds also see option -ddcheck
>>
>>
>> ######################################################################
>>
>> How to solve this error? Is there any way to customize the 'gmx mdrun' command inside the scripts?
>>
>>
>> Another issue: How to include exclusions in topol.top?
>> (I have created the file exclusions.txt, and included it in topol.top. But, csg_inverse cannot find it)
>>
>>
>> Thanks in advance!
>>
>> -Souvik
>>
> --
> You received this message because you are subscribed to the Google Groups "votca" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to votca+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/a4d05a4f-e028-473e-b0d5-e86edef176b8%40googlegroups.com.

Souvik Chakraborty

unread,
Aug 17, 2019, 5:08:07 AM8/17/19
to votca
Thanks, Christoph.

Before csg_inverse, the force is still very high even after several em steps on conf_cg.gro. But, my atomistic trajectory was well-equilibrated.

I am not not sure how to get the force within tolerance for a CG geometry. Should I equilibrate the atomistic trajectory more, and start from scratch?

Regards,
Souvik

On Wednesday, 14 August 2019 20:40:19 UTC+8, Souvik Chakraborty wrote:

Christoph Junghans

unread,
Aug 17, 2019, 9:12:55 AM8/17/19
to vo...@googlegroups.com
On Sat, Aug 17, 2019 at 3:08 AM Souvik Chakraborty <souv...@gmail.com> wrote:
>
> Thanks, Christoph.
>
> Before csg_inverse, the force is still very high even after several em steps on conf_cg.gro. But, my atomistic trajectory was well-equilibrated.
>
> I am not not sure how to get the force within tolerance for a CG geometry. Should I equilibrate the atomistic trajectory more, and start from scratch?
This really depends on the system, but I would have a look at the
table in step_001 and see if they look ok to start with. Usually the
initial condition, which is what you get from mapping the atomistic
trajectory, isn't super important, but it might be worth minimizing it
with just repulsive lennard-jones.

Christoph

>
> Regards,
> Souvik
>
> On Wednesday, 14 August 2019 20:40:19 UTC+8, Souvik Chakraborty wrote:
>>
>> Dear group members,
>>
>> While running csg_inverse to get non-bonded potential, I get the following error in inverse.log:
>>
>> Fatal error:
>> 1 of the 4500 bonded interactions could not be calculated because some atoms
>> involved moved further apart than the multi-body cut-off distance (1.5 nm) or
>> the two-body cut-off distance (1.5 nm), see option -rdd, for pairs and
>> tabulated bonds also see option -ddcheck
>>
>>
>> ######################################################################
>>
>> How to solve this error? Is there any way to customize the 'gmx mdrun' command inside the scripts?
>>
>>
>> Another issue: How to include exclusions in topol.top?
>> (I have created the file exclusions.txt, and included it in topol.top. But, csg_inverse cannot find it)
>>
>>
>> Thanks in advance!
>>
>> -Souvik
>>
> --
> You received this message because you are subscribed to the Google Groups "votca" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to votca+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/4c072f72-e174-4fc1-8e36-fe4ce9b65092%40googlegroups.com.

Souvik Chakraborty

unread,
Aug 19, 2019, 7:59:47 AM8/19/19
to votca
Thanks, Christoph for the suggestions.

I will try to sort this out - with and without fmatch output!

Kind regards,
Souvik

On Wednesday, 14 August 2019 20:40:19 UTC+8, Souvik Chakraborty wrote:
Reply all
Reply to author
Forward
0 new messages