IBI on a polymer melt

124 views
Skip to first unread message

CHRISTINA-ANNA GATSIOU

unread,
Dec 13, 2019, 5:35:26 PM12/13/19
to votca
Dear Christoph, 

I am working on a polymer melt and I am consistently getting rdfs for nonbonded interactions that look like the attachment. I get a peak where there is a bond and it shouldn't be counted as a non bonded interaction. I saw in earlier posts that one should first derive the bonded interactions from a single chain and then nonbonded from the melt. 
Even though I did not follow this strategy still I wouldn't expect something like this. Any help would be very much appreciated. 

Best regards, 
Christianna



P2_P2.eps

Christoph Junghans

unread,
Dec 13, 2019, 6:26:09 PM12/13/19
to vo...@googlegroups.com
it looks to me like something is wrong with your exclusions, i.e. some
internal bonds aren't excluded, have a look at
$ csg_dump --excl --top <topology_file>

You might need to some bonds to your topology, which will
automatically generate exclusions for you.

Christoph
>
> Best regards,
> Christianna
>
>
>
> --
> 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/baabfba8-b22d-475c-a272-99195327842e%40googlegroups.com.



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

CHRISTINA-ANNA GATSIOU

unread,
Dec 14, 2019, 1:27:54 PM12/14/19
to vo...@googlegroups.com
Hi Christoph, 

My topology.xml does contain exclusions. I have only specified nrexcl 3 which I thought should do the job. Do I need to specify exclusions in the topology.xml or in the gromacs topology file ?
Thanks for the prompt response. 

Best regards, 
Christianna 
On 14 Dec 2019, at 1:26 AM, Christoph Junghans <jung...@votca.org> wrote:

csg_dump --excl --top <topology_file>

Christoph Junghans

unread,
Dec 14, 2019, 1:52:34 PM12/14/19
to vo...@googlegroups.com
On Sat, Dec 14, 2019 at 11:27 CHRISTINA-ANNA GATSIOU <christian...@gmail.com> wrote:
Hi Christoph, 

My topology.xml does contain exclusions. I have only specified nrexcl 3 which I thought should do the job. Do I need to specify exclusions in the topology.xml or in the gromacs topology file ?
If you are using a tpr file as topology VOTCA will read the exclusions from there. If you are using a xml topology exclusions will be generated out of the bonds and angles.

Just look in inverse.log what topology file is used by csg_stat and run csg_dump on it to check the exclusions.

Christoph 
Thanks for the prompt response. 

Best regards, 
Christianna 
On 14 Dec 2019, at 1:26 AM, Christoph Junghans <jung...@votca.org> wrote:

csg_dump --excl --top <topology_file>

--
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.

CHRISTINA-ANNA GATSIOU

unread,
Jan 3, 2020, 10:13:22 AM1/3/20
to votca
Hi Christoph,

I have set all exclusions in the gromacs topology file. However, during IBI for one nonbonded interaction I get the attached distribution (red corresponds to the bond distribution the bonded beads, green corresponds to the target distribution of bead types and blue to IBI distribution at step 30). It seems that bonded pairs are included in the g(r) even though I have specified them as non interacting in the exclusion list. I cannot figure out what might be the problem here. Thanks in advance.

 
Best regards,
Christianna

On Saturday, December 14, 2019 at 8:52:34 PM UTC+2, Christoph Junghans wrote:
On Sat, Dec 14, 2019 at 11:27 CHRISTINA-ANNA GATSIOU <christian...@gmail.com> wrote:
Hi Christoph, 

My topology.xml does contain exclusions. I have only specified nrexcl 3 which I thought should do the job. Do I need to specify exclusions in the topology.xml or in the gromacs topology file ?
If you are using a tpr file as topology VOTCA will read the exclusions from there. If you are using a xml topology exclusions will be generated out of the bonds and angles.

Just look in inverse.log what topology file is used by csg_stat and run csg_dump on it to check the exclusions.

Christoph 
Thanks for the prompt response. 

Best regards, 
Christianna 
On 14 Dec 2019, at 1:26 AM, Christoph Junghans <jung...@votca.org> wrote:

csg_dump --excl --top <topology_file>

--
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 vo...@googlegroups.com.

Christoph Junghans

unread,
Jan 3, 2020, 10:30:28 AM1/3/20
to vo...@googlegroups.com
On Fri, Jan 3, 2020 at 8:13 AM CHRISTINA-ANNA GATSIOU <christian...@gmail.com> wrote:
Hi Christoph,

I have set all exclusions in the gromacs topology file. However, during IBI for one nonbonded interaction I get the attached distribution (red corresponds to the bond distribution the bonded beads, green corresponds to the target distribution of bead types and blue to IBI distribution at step 30). It seems that bonded pairs are included in the g(r) even though I have specified them as non interacting in the exclusion list. I cannot figure out what might be the problem here. Thanks in advance.
Go in the log file (usually inverse.log) and find the last csg_stat call (e.g. with "grep csg_stat inverse.log") and post it here.

Christoph

 
Best regards,
Christianna

On Saturday, December 14, 2019 at 8:52:34 PM UTC+2, Christoph Junghans wrote:


On Sat, Dec 14, 2019 at 11:27 CHRISTINA-ANNA GATSIOU <christian...@gmail.com> wrote:
Hi Christoph, 

My topology.xml does contain exclusions. I have only specified nrexcl 3 which I thought should do the job. Do I need to specify exclusions in the topology.xml or in the gromacs topology file ?
If you are using a tpr file as topology VOTCA will read the exclusions from there. If you are using a xml topology exclusions will be generated out of the bonds and angles.

Just look in inverse.log what topology file is used by csg_stat and run csg_dump on it to check the exclusions.

Christoph 
Thanks for the prompt response. 

Best regards, 
Christianna 
On 14 Dec 2019, at 1:26 AM, Christoph Junghans <jung...@votca.org> wrote:

csg_dump --excl --top <topology_file>

--
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 vo...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/votca/7889732D-B264-4FBB-A951-271C1190FB7F%40gmail.com.
--
Christoph Junghans
Web: http://www.compphys.de

--
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/c927b5cc-550d-4388-b10f-ae598ec15753%40googlegroups.com.

CHRISTINA-ANNA GATSIOU

unread,
Jan 6, 2020, 12:26:36 PM1/6/20
to votca


On Friday, January 3, 2020 at 5:30:28 PM UTC+2, Christoph Junghans wrote:


On Fri, Jan 3, 2020 at 8:13 AM CHRISTINA-ANNA GATSIOU <christian...@gmail.com> wrote:
Hi Christoph,

I have set all exclusions in the gromacs topology file. However, during IBI for one nonbonded interaction I get the attached distribution (red corresponds to the bond distribution the bonded beads, green corresponds to the target distribution of bead types and blue to IBI distribution at step 30). It seems that bonded pairs are included in the g(r) even though I have specified them as non interacting in the exclusion list. I cannot figure out what might be the problem here. Thanks in advance.
Go in the log file (usually inverse.log) and find the last csg_stat call (e.g. with "grep csg_stat inverse.log") and post it here.

csg_stat --nt 8 --options settings_nb.xml --top topol.tpr --trj traj.xtc --begin 5 --first-frame 0 --cg topology.xml;

 
Christoph

Christoph Junghans

unread,
Jan 6, 2020, 1:34:01 PM1/6/20
to vo...@googlegroups.com
On Mon, Jan 6, 2020 at 10:26 AM CHRISTINA-ANNA GATSIOU
<christian...@gmail.com> wrote:
>
>
>
> On Friday, January 3, 2020 at 5:30:28 PM UTC+2, Christoph Junghans wrote:
>>
>>
>>
>> On Fri, Jan 3, 2020 at 8:13 AM CHRISTINA-ANNA GATSIOU <christian...@gmail.com> wrote:
>>>
>>> Hi Christoph,
>>>
>>> I have set all exclusions in the gromacs topology file. However, during IBI for one nonbonded interaction I get the attached distribution (red corresponds to the bond distribution the bonded beads, green corresponds to the target distribution of bead types and blue to IBI distribution at step 30). It seems that bonded pairs are included in the g(r) even though I have specified them as non interacting in the exclusion list. I cannot figure out what might be the problem here. Thanks in advance.
>>
>> Go in the log file (usually inverse.log) and find the last csg_stat call (e.g. with "grep csg_stat inverse.log") and post it here.
>>
> csg_stat --nt 8 --options settings_nb.xml --top topol.tpr --trj traj.xtc --begin 5 --first-frame 0 --cg topology.xml;
Ok, if you are using a mapping file (--cg option), the exclusion from
the tpr get mapped out. You will need to add bonds between the atoms
you want to exclude in mapping file. See here for an example:
https://github.com/votca/csg-tutorials/blob/master/hexane/ibi_all/hexane_cg.xml#L25-L39
You don't need to do anything with these bonds in the IBI iterations.

Again you can check the exclusions after mapping with csg_dump, e.g.
$ csg_dump --excl --top topol.tpr --cg topology.xml

Christoph
>
>
>>
>> Christoph
> 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/92a68fed-7eca-41d6-90cb-f7825f229c82%40googlegroups.com.

CHRISTINA-ANNA GATSIOU

unread,
Jan 8, 2020, 4:38:08 AM1/8/20
to votca
> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/92a68fed-7eca-41d6-90cb-f7825f229c82%40googlegroups.com.


I already use a mapping file with bonds defined between these beads.

CHRISTINA-ANNA GATSIOU

unread,
Jan 8, 2020, 4:56:53 AM1/8/20
to votca
Hi Christoph,

"Ok, if you are using a mapping file (--cg option), the exclusion from the tpr get mapped out."
Does this mean that I should not be using the mapping file in order for the exclusions to be taken into account. I was under the impression that it was necessary. I have already defined those bonds in the mapping file.

Christoph Junghans

unread,
Jan 8, 2020, 8:10:37 AM1/8/20
to vo...@googlegroups.com
On Wed, Jan 8, 2020 at 02:56 CHRISTINA-ANNA GATSIOU <christian...@gmail.com> wrote:
Hi Christoph,

"Ok, if you are using a mapping file (--cg option), the exclusion from the tpr get mapped out."
Does this mean that I should not be using the mapping file in order for the exclusions to be taken into account. I was under the impression that it was necessary. I have already defined those bonds in the mapping file.
It depends, a mapping file is usually only needed if you want to do IBI on the bonded interactions as well.

As VOTCA has no concept of nrexcl in gromacs once you map, you might need to more unphysical bonds, e.g. using a tpr with a chain with bonds A-B-C, A & C would be excluded if nrexcl is 2, after mapping A & B as well as B & C are excluded, but A & C are not, so you will need to add an unphysical bond between A & C to achieve that. Looking at the output of “csg_dump —excl —top topol” vs. “csg_dump —excl —top topol.tpr —cg map.xml” should reveal that.

Can you send your tpr and mapping files?
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/c352c13c-4676-4b79-a75b-ddaa44ab9919%40googlegroups.com.

CHRISTINA-ANNA GATSIOU

unread,
Jan 16, 2020, 10:38:06 AM1/16/20
to votca
Hi Christoph,

I managed to resolve the issue with non-bonded interactions by not using the toopology.xml when fitting the non-bonded interactions and by using nrexcl is 3 in topol.top.
However it does not make sense why it did not work before since bonds etc were appropriately specified.
There is one additional problem with the convergence of angles and dihedrals. Having fixed the nonbonded interactions, the angle distributions do not seem to improve.
 
Best regards,
Christianna
topology_.xml
B2-B4-B2.eps
topol.tpr

Christoph Junghans

unread,
Jan 16, 2020, 11:29:32 AM1/16/20
to vo...@googlegroups.com
On Thu, Jan 16, 2020 at 8:38 AM CHRISTINA-ANNA GATSIOU <christian...@gmail.com> wrote:
Hi Christoph,

I managed to resolve the issue with non-bonded interactions by not using the toopology.xml when fitting the non-bonded interactions and by using nrexcl is 3 in topol.top.
However it does not make sense why it did not work before since bonds etc were appropriately specified.
I looked at topology_.xml brief and it is really long, so I have two guesses:
- you simply forgot to list a bond
- nrexcl corresponds to 3 bond length exclusions, .i.e. in a bond A-B-C-D, A and D are excluded, even though there is no bond between A and D, so you would need to add a non-physical bond to your topology_.xml  
There is one additional problem with the convergence of angles and dihedrals. Having fixed the nonbonded interactions, the angle distributions do not seem to improve.
 dihedrals have long convergence in general, but angles could be slow as well depending on your system. Did you look at the angle auto-correlation time to get a feel if the distribution are sampled went enough?

Christoph
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/dba049a5-ec66-49d0-9a7b-122b13fc9dbc%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages