Error in using csg_map with a pdb file

[Rishabh Guha]

Hello,

I have an atomistic system consisting of 250 alanine and 250 valine molecules. I am trying to use cg_map using the pdb file and the LAMMPS output trajectory, but I am getting an error:-

cannot find: <1:VAL:CA> in VAL

an error occurred:

mapping error: molecule 1:VAL:CA does not exist

My topology file looks like:-

<topology base="temp.pdb">

<molecules>

<clear/>

<define name="ALA" first="1" nbeads="13" nmols="250"/>

<define name="VAL" first="3251" nbeads="19" nmols="250"/>

</molecules>

</topology>

My individual alanine and valine xml files are:-

Alanine-

<cg_molecule>

  <name>ALA</name>

  <ident>ALA</ident>

  <topology>

    <cg_beads>

      <cg_bead>

        <name>ALA</name>

        <type>ALA</type>

        <mapping>A</mapping>

        <beads>

           1:ALA:CA 1:ALA:C 1:ALA:N 1:ALA:H1 1:ALA:H2 1:ALA:CB 1:ALA:HA 1:ALA:HB1 1:ALA:HB2 1:ALA:HB3 1:ALA:O 1:ALA:OXT 1:ALA:HXT

        </beads>

      </cg_bead>

  </cg_beads>

  </topology>

  <maps>

    <map>

      <name>A</name>

      <weights>12 12 14 1 1 12 1 1 1 1 16 16 1</weights>

    </map>

  </maps>

</cg_molecule>

Valine 

<cg_molecule>

  <name>VAL</name>

  <ident>VAL</ident>

  <topology>

    <cg_beads>

      <cg_bead>

        <name>VAL</name>

        <type>VAL</type>

        <mapping>A</mapping>

        <beads>

           1:VAL:CA 1:VAL:C 1:VAL:N 1:VAL:H1 1:VAL:H2 1:VAL:CB 1:VAL:HA 1:VAL:CG1 1:VAL:HG11 1:VAL:HG12 1:VAL:HG13 1:VAL:CG2 1:VAL:HB 1:VAL:HG21 1:VAL:HG22 1:VAL:HG23 1:VAL:O 1:VAL:OXT 1:VAL:HXT

        </beads>

      </cg_bead>

  </cg_beads>

  </topology>

  <maps>

    <map>

      <name>A</name>

      <weights>12 12 14 1 1 12 1 12 1 1 1 12 1 1 1 1 16 16 1</weights>

    </map>

  </maps>

</cg_molecule>

I have attached the pdb file for further reference.

Can anyone please help me with this?

[Christoph Junghans]

On Wed, Jul 7, 2021 at 13:23 Rishabh Guha <rdg...@ncsu.edu> wrote:

Hello,

I have an atomistic system consisting of 250 alanine and 250 valine molecules. I am trying to use cg_map using the pdb file and the LAMMPS output trajectory, but I am getting an error:-

cannot find: <1:VAL:CA> in VAL

an error occurred:

mapping error: molecule 1:VAL:CA does not exist

try to run “csg_dump --top topol.xml” and check how VOTCA labels things.

Christoph 

--
Join us on Slack: https://join.slack.com/t/votca/signup
---
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/3c0b76f3-ff3a-4013-8b74-d366e1e9c865n%40googlegroups.com.

--

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

[Rishabh Guha]

Hey Christoph,

Thanks for the help. I got to know about the csg_dump command as soon as I sent you the email. For some reason, csg_dump is categorizing all the atoms as the valine molecules as ALA as well, which is counterintuitive to me. But I did change my valine.xml file to:-

<cg_molecule>
  <name>VAL</name>
  <ident>VAL</ident>
  <topology>
    <cg_beads>
      <cg_bead>
        <name>VAL</name>
        <type>VAL</type>
        <mapping>A</mapping>
        <beads>

           1:ALA:CA 1:ALA:C 1:ALA:N 1:ALA:H1 1:ALA:H2 1:ALA:CB 1:ALA:HA 1:ALA:CG1 1:ALA:HG11 1:ALA:HG12 1:ALA:HG13 1:ALA:CG2 1:ALA:HB 1:ALA:HG21 1:ALA:HG22 1:ALA:HG23 1:ALA:O 1:ALA:OXT 1:ALA:HXT

        </beads>
      </cg_bead>
  </cg_beads>
  </topology>
  <maps>
    <map>
      <name>A</name>

      <weights>12 12 14 1 1 12 1 12 1 1 1 12 1 1 1 1 16 16 1</weights>
    </map>
  </maps>
</cg_molecule>

Now, I get a different error,

I have 8000 beads in 500 molecules
I have 500 beads in 500 molecules for the coarsegraining
 1.1853
-0.2654
 0.0787  1.2896
-0.2967
-0.0276
an error occurred:
coarse-grained bead is bigger than half the box
 (atoms CA (id 1), C (id 2) , molecule 1)

Can you please tell me what I can change?

You received this message because you are subscribed to a topic in the Google Groups "votca" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/votca/C3L0xhdUs6s/unsubscribe.
To unsubscribe from this group and all its topics, send an email to votca+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e7WZfOD2xdHy1AWWob7zNtdr5cBpYydNGp0WmA9sfqO9w%40mail.gmail.com.

--

Rishabh Debraj Guha(PhD student)

Graduate Research Assistant

Mechanical and Aerospace Engineering

North Carolina State University

Contact:- +1-347-205-2280

[Christoph Junghans]

On Wed, Jul 7, 2021 at 13:44 Rishabh Guha <rdg...@ncsu.edu> wrote:

Hey Christoph,

Thanks for the help. I got to know about the csg_dump command as soon as I sent you the email. For some reason, csg_dump is categorizing all the atoms as the valine molecules as ALA as well, which is counterintuitive to me.

yeah, there is a way to fix this, these are just unique identifiers so it doesn’t matter a lot what they are as long as they are unique. This artifact mainly stems from the fact that the pdb format misses some molecule information.

this error means you are trying to map teo atoms that are more than half a box length away into one cg bead. The 6 number above are the positions.

There can be multiple reasons for that error, most of the times it is a typo in the mapping file. If you are sure things are correct, you can always make the box bigger.

Christoph 

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktCXqB_zefYDpusjSBnLkt%2B77vfkgnU-dovocsFsiiKUow%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

Thanks for the pointers. I tried to look closely into my mapping files and compared them with the pdb files, but nothing jumped out at me. Although I admit that it can easily be an oversight-I have just started using Votca today. I have attached all my files below. Does anything jump out at you? I am convinced that none of my atoms are than half the box length away

Regards

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e6Kjvwxvv4suXuuqLSQ57AJeh_Vi-CepLi4A%3DXb9MmzWg%40mail.gmail.com.

[Christoph Junghans]

On Wed, Jul 7, 2021, 14:05 Rishabh Guha <rdg...@ncsu.edu> wrote:

Hey Christoph,

Thanks for the pointers. I tried to look closely into my mapping files and compared them with the pdb files, but nothing jumped out at me. Although I admit that it can easily be an oversight-I have just started using Votca today. I have attached all my files below. Does anything jump out at you? I am convinced that none of my atoms are than half the box length away

The error says there is an atom at 1.1853

-0.2654 0.0787  and one at 1.2896 -0.2967 -0.0276 that you are trying to map in a cg bead. These are the ones you have to look at carefully in the mapping file.

Christoph

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktBEGaPFAtudiU20y7Y-0pwe1jS2XXnvShd0kyXvfBLuOQ%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

First of all in my pdb file the atoms are at:-

ATOM      1  CA  ALA A   1      11.853  -2.654   0.787  1.00  0.00           C  
ATOM      2  C   ALA A   1      12.896  -2.967  -0.276  1.00  0.00           C

Do you know why votca is dividing all the positions by 10?

Secondly these atoms show up as- 

0 Name 1:ALA:CA Type CA Mass 12.0107 Resnr 0 Resname ALA Charge 0
1 Name 1:ALA:C Type C Mass 12.0107 Resnr 0 Resname ALA Charge 0

when I do csg_dump --top topology.xml and are mapped as 1:ALA:CA 1:ALA:C 1:ALA:N 1:ALA:H1 1:ALA:H2 1:ALA:CB 1:ALA:HA 1:ALA:HB1 1:ALA:HB2 1:ALA:HB3 1:ALA:O 1:ALA:OXT 1:ALA:HXT in the alanine.xml file.

I do not see any issues in the mapping to be honest

Regards

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e4ypf5Po8_zKmhBSbSTW_xgGBu09Bmj6Nsh8KdQtYJN-Q%40mail.gmail.com.

[Christoph Junghans]

On Wed, Jul 7, 2021 at 2:18 PM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

> First of all in my pdb file the atoms are at:-
> ATOM 1 CA ALA A 1 11.853 -2.654 0.787 1.00 0.00 C
> ATOM 2 C ALA A 1 12.896 -2.967 -0.276 1.00 0.00 C
>
> Do you know why votca is dividing all the positions by 10?

VOTCA uses nm internally, and pdb is in Angstrom.

>
> Secondly these atoms show up as-
> 0 Name 1:ALA:CA Type CA Mass 12.0107 Resnr 0 Resname ALA Charge 0
> 1 Name 1:ALA:C Type C Mass 12.0107 Resnr 0 Resname ALA Charge 0
> when I do csg_dump --top topology.xml and are mapped as 1:ALA:CA 1:ALA:C 1:ALA:N 1:ALA:H1 1:ALA:H2 1:ALA:CB 1:ALA:HA 1:ALA:HB1 1:ALA:HB2 1:ALA:HB3 1:ALA:O 1:ALA:OXT 1:ALA:HXT in the alanine.xml file.
>
> I do not see any issues in the mapping to be honest

When you run csg_dump, what does it tell you the box size is?

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktBH-gARMiCQR%2BdJVudO95mHB5ZZkf6odZJDxmp6U2a-nQ%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

First of all thanks for following up. Sorry this is gonna be a long email but I have some interesting observations based on your question. First of all if I try to do a topology dump using my pdb file I don't get a "size" of the box per say. 

I get this :- I have 8000 beads in 500 molecules

Boundary Condition: open

Moreover the charges on my atoms are also 0 in the pdb file. I figured that for my case, using the LAMMPS data file I generate from my PDB file is a better candidate. When I use the LAMMPS data file, the identifiers change from 1:ALA:XX to 1:DUM:XX. I did all those changes and then if I do csg_dump I get:-

I have 11000 beads in 500 molecules
Boundary Condition: orthorhombic
 Box matix: 5.76509 0 0
            0 5.76509 0
            0 0 5.76509 (It was a different data set which has 11000 molecules instead of 8000 so that is not an issue)

Now my original LAMMPS data file actually gies from:

-28.825434 28.825434 xlo xhi
-28.825434 28.825434 ylo yhi
-28.825434 28.825434 zlo zhi

So I guess VOTCA doesn't account for the negative coordinates and starts everything from 0.

I tried csg_map with this and I again got the previous error:

I have 11000 beads in 500 molecules

I have 500 beads in 500 molecules for the coarsegraining

-20.153
  0.534
-24.577  -18.68
  3.621
-23.916

an error occurred:
coarse-grained bead is bigger than half the box

 (atoms N1 (id 18), O7 (id 22) , molecule 1).

This time around I just increased the box size by a multiplier of 10 in x,y and z and it worked:

These styles use the following formats in the atom block:
atom-ID molecule-ID atom-type charge x y z
atom-ID molecule-ID atom-type charge x y z nx ny nz
atom-ID molecule-ID atom-type x y z
atom-ID molecule-ID atom-type x y z nx ny nz
atom-ID atom-type x y z
atom-ID atom-type x y z nx ny nz

I have 11000 beads in 500 molecules

I have 500 beads in 500 molecules for the coarsegraining

Reading frame, timestep 5000000
writing coarse-grained trajectory to cg.gro

I wanted to know whether increasing the box size is okay in this case. 

Thanks again

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e58qQRPUVC07BnJTeMtkm5StheQQTxsSQsDo_fOQz8Lcw%40mail.gmail.com.

[Christoph Junghans]

On Thu, Jul 8, 2021 at 8:21 AM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

> First of all thanks for following up. Sorry this is gonna be a long email but I have some interesting observations based on your question. First of all if I try to do a topology dump using my pdb file I don't get a "size" of the box per say.
>
> I get this :- I have 8000 beads in 500 molecules
> Boundary Condition: open
>
> Moreover the charges on my atoms are also 0 in the pdb file. I figured that for my case, using the LAMMPS data file I generate from my PDB file is a better candidate. When I use the LAMMPS data file, the identifiers change from 1:ALA:XX to 1:DUM:XX. I did all those changes and then if I do csg_dump I get:-
>
> I have 11000 beads in 500 molecules
> Boundary Condition: orthorhombic
> Box matix: 5.76509 0 0
> 0 5.76509 0
> 0 0 5.76509 (It was a different data set which has 11000 molecules instead of 8000 so that is not an issue)
>
> Now my original LAMMPS data file actually gies from:
> -28.825434 28.825434 xlo xhi
> -28.825434 28.825434 ylo yhi
> -28.825434 28.825434 zlo zhi
>
> So I guess VOTCA doesn't account for the negative coordinates and starts everything from 0.

Well, 2x 28.825434 Angstroem = 5.76509 nm, so above it is correct.
VOTCA uses minimum image convention when a box is given, so it is able
deal with negative coordinates.

>
> I tried csg_map with this and I again got the previous error:
> I have 11000 beads in 500 molecules
> I have 500 beads in 500 molecules for the coarsegraining
> -20.153
> 0.534
> -24.577 -18.68
> 3.621
> -23.916
> an error occurred:
> coarse-grained bead is bigger than half the box
> (atoms N1 (id 18), O7 (id 22) , molecule 1).

Half box length 2.88254 nm, but | 3.621 - 0.534 | (y difference of the
two beads above) is 3.087nm and hence bigger than L/2, so the error
message is correct as well.

>
> This time around I just increased the box size by a multiplier of 10 in x,y and z and it worked:
>
> These styles use the following formats in the atom block:
> atom-ID molecule-ID atom-type charge x y z
> atom-ID molecule-ID atom-type charge x y z nx ny nz
> atom-ID molecule-ID atom-type x y z
> atom-ID molecule-ID atom-type x y z nx ny nz
> atom-ID atom-type x y z
> atom-ID atom-type x y z nx ny nz
>
> I have 11000 beads in 500 molecules
> I have 500 beads in 500 molecules for the coarsegraining
> Reading frame, timestep 5000000
> writing coarse-grained trajectory to cg.gro
>
> I wanted to know whether increasing the box size is okay in this case.

When you increase the box size you basically force the atoms in the
1st image of the periodic box.
One problem with this method is that one could construct a case where
a molecule would be broken into 2 pieces, when some atoms have
coordinates on a different image than other atoms. Only way to know is
to look at the output with vmd or pymol.

Christoph

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktD6b4Tqy%3DMAY6bS1BFwiSiXjtpYeGZ0ffg7Y3QAVLjaXA%40mail.gmail.com.

[Rishabh Guha]

I think there is an inconsistency here because these coordinates are straight from my LAMMPS data file:-

I have 500 beads in 500 molecules for the coarsegraining
> -20.153
>   0.534
> -24.577  -18.68
>   3.621
> -23.916
> an error occurred:
> coarse-grained bead is bigger than half the box
>  (atoms N1 (id 18), O7 (id 22) , molecule 1).

18 1 1 -0.9187999963760376 -20.153000 0.534000 -24.577000

22 1 7 -0.4959999918937683 -18.680000 3.621000 -23.916000

Therefore, these are in Angstrom (not nm). So 3.621 - 0.534 is technically 0.3087 nm. Not sure why Votca is not doing the nm conversion here. 

Anyways, I tried visualizing the .gro file in VMD and it looks fine. Nothing stands out. The initial file was 250 molecules of 2 different types of amino acids. Since I mapped each molecule to a single bead I now get 2 different types of beads each being 250 in number.

Can you please give me some pointers on how I proceed now. My aim is to generate a CG potential for these amino acid pairs based on RDF calculated from the atomistic trajectories. I have already calculated the RDF's from the atomistic trajectory but I don't know how I use it in VOTCA. 

I was trying to follow some discussion topics on the group and one of them mentions generating a CG topology file using cg_gmxtopol. When I try that I am getting this warning :-

csg_gmxtopol --top topology.xml --cg "lysine.xml;methionine.xml" --out topol.top

WARNING: The votca lammps data reader is only able to read lammps files formatted in the following styles:
angle
atom
bond
full
molecule

These styles use the following formats in the atom block:
atom-ID molecule-ID atom-type charge x y z
atom-ID molecule-ID atom-type charge x y z nx ny nz
atom-ID molecule-ID atom-type x y z
atom-ID molecule-ID atom-type x y z nx ny nz
atom-ID atom-type x y z
atom-ID atom-type x y z nx ny nz

I have 11000 beads in 500 molecules
I have 500 beads in 500 molecules for the coarsegraining

WARNING: cannot create topology for topology withmultiple molecules, using only first molecule

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e7OFP%2B_yTA0BNk4q07NbGfdH29QOpi0PoQ%3DdOPon2kTYA%40mail.gmail.com.

[Christoph Junghans]

On Thu, Jul 8, 2021 at 9:19 AM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> I think there is an inconsistency here because these coordinates are straight from my LAMMPS data file:-
> I have 500 beads in 500 molecules for the coarsegraining
> > -20.153
> > 0.534
> > -24.577 -18.68
> > 3.621
> > -23.916
> > an error occurred:
> > coarse-grained bead is bigger than half the box
> > (atoms N1 (id 18), O7 (id 22) , molecule 1).
>
> 18 1 1 -0.9187999963760376 -20.153000 0.534000 -24.577000
> 22 1 7 -0.4959999918937683 -18.680000 3.621000 -23.916000
>
> Therefore, these are in Angstrom (not nm). So 3.621 - 0.534 is technically 0.3087 nm. Not sure why Votca is not doing the nm conversion here.

You are right, I made a fix here:
https://github.com/votca/csg/pull/697. We fixed it in the dump reader
before, but seem to have missed it in the LAMMPS data reader.

>
> Anyways, I tried visualizing the .gro file in VMD and it looks fine. Nothing stands out. The initial file was 250 molecules of 2 different types of amino acids. Since I mapped each molecule to a single bead I now get 2 different types of beads each being 250 in number.
>
> Can you please give me some pointers on how I proceed now. My aim is to generate a CG potential for these amino acid pairs based on RDF calculated from the atomistic trajectories. I have already calculated the RDF's from the atomistic trajectory but I don't know how I use it in VOTCA.

The csg manual has more details here: https://www.votca.org/csg/index.html
But in short, you will have to generate input files for Gromacs (or
LAMMPS) that have everything but the potentials.
csg_gmxtopol can be a good starting point even though it can only deal
with one type of molecule, but I usually start with one of the
tutorials go from there, if you have 2 kinds of molecules, have a look
at the methanol-water or urea-water tutorials:
https://github.com/votca/csg-tutorials

Christoph

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktB2v_gQSQxrnuwOk382V-N6FustzWs%3DoP8%3DY-A_0n2WRg%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

Thanks for the input. I was going over the tutorials and I think in my case, the ibi_lammps should work if I just change the .in.data and CG-CG.dist.tgt file. When I try to run this (even with the default spce.in and spce.data file), I get this error:

sh run.sh
running csg_inverse --options "settings.xml"
Appending to existing logfile inverse.log
We are doing Method: ibi
Prepare (dir step_000)
Using initial guess from dist CG-CG.dist.tgt for CG-CG
step 0 done
Doing iteration 1 (dir step_001)
Simulation with lammps
################################################################################################################################################
#                                                                                                                                              #
# ERROR:                                                                                                                                       #
# Command/function lmp not found (when calling from csg_call you might need to add --simprog option or set cg.inverse.program in the xml file) #
# For details see the logfile /ccs/home/rdguha/votca/csg-tutorials/spce/ibi_lammps/inverse.log                                                 #
#                                                                                                                                              #
################################################################################################################################################
run.sh: line 4: 2698557 Terminated              csg_inverse --options settings.xml

I have attached the settings.xml file with this email. Do you know what I have to change and where for lammps to work? 

Thanks

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e5AEia5NhbG3BPfb27Z2GTySAqoWR5BP%2BZX5edvimmjAQ%40mail.gmail.com.

[Christoph Junghans]

On Thu, Jul 8, 2021 at 11:04 AM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

> Thanks for the input. I was going over the tutorials and I think in my case, the ibi_lammps should work if I just change the .in.data and CG-CG.dist.tgt file. When I try to run this (even with the default spce.in and spce.data file), I get this error:
> sh run.sh
> running csg_inverse --options "settings.xml"
> Appending to existing logfile inverse.log
> We are doing Method: ibi
> Prepare (dir step_000)
> Using initial guess from dist CG-CG.dist.tgt for CG-CG
> step 0 done
> Doing iteration 1 (dir step_001)
> Simulation with lammps
> ################################################################################################################################################
> # #
> # ERROR: #
> # Command/function lmp not found (when calling from csg_call you might need to add --simprog option or set cg.inverse.program in the xml file) #
> # For details see the logfile /ccs/home/rdguha/votca/csg-tutorials/spce/ibi_lammps/inverse.log #
> # #
> ################################################################################################################################################
> run.sh: line 4: 2698557 Terminated csg_inverse --options settings.xml
>
> I have attached the settings.xml file with this email. Do you know what I have to change and where for lammps to work?

The error says it all, the lammps executable ("lmp") was not found,
hence it stopped.

You can change the lammps default name at cmake time using the
LMP_EXECUTABLE option.
Alternatively you can set the cg.inverse.lammps.command in your xml file.

Christoph

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktB25AKd9yfZ%3DfHDKDJA0ghNJwkhbmPuBK%3DDCz84tzwUdA%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

I appreciate all the help but there is still a small roadblock I am facing. 

The xml file as you know has this setting by default :-

<!-- general options for inverse script -->
  <inverse>
    <!-- votca units 300*0.00831451 kJ/mol -->
    <kBT>2.49435</kBT>
    <!-- use lammps as simulation program -->
    <program> lammps </program>
    <!-- lammps specific options -->
    <lammps>
      <!-- lammps script to run !-->
      <script>spce.in</script>
      <!-- topology to be used by  csg_stat !-->
      <topol>spce.data</topol>
      <!-- traj file created by lammps !-->
      <traj>traj.dump</traj>
    </lammps>
    <initial_configuration>maindir</initial_configuration>
    <!-- these files are copied for each new run -->
    <filelist>spce.data spce.in</filelist>
    <!-- do so many iterations -->
    <iterations_max>300</iterations_max>
    <!-- ibi: inverse biltzmann imc: inverse monte carlo -->
    <method>ibi</method>
  </inverse>

I changed it to :-

<!-- general options for inverse script -->
  <inverse>
    <!-- votca units 300*0.00831451 kJ/mol -->
    <kBT>2.49435</kBT>
    <!-- use lammps as simulation program -->
    <program> /gpfs/alpine/mat222/proj-shared/rdguha/lammps/build/lmp (path to my lammps executable) <program>
    <!-- lammps specific options -->
    <lammps>
      <!-- lammps script to run !-->
      <script>spce.in</script>
      <!-- topology to be used by  csg_stat !-->
      <topol>spce.data</topol>
      <!-- traj file created by lammps !-->
      <traj>traj.dump</traj>
    </lammps>
    <initial_configuration>maindir</initial_configuration>
    <!-- these files are copied for each new run -->
    <filelist>spce.data spce.in</filelist>
    <!-- do so many iterations -->
    <iterations_max>300</iterations_max>
    <!-- ibi: inverse biltzmann imc: inverse monte carlo -->
    <method>ibi</method>
  </inverse>

But now, I get this error :-

For a more verbose log see: inverse.log

We are doing Method: ibi

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

#                                                                                                                          #
# ERROR:                                                                                                                   #

# source_wrapper: Could not get any script from tags 'functions' '/gpfs/alpine/mat222/proj-shared/rdguha/lammps/build/lmp' #
# For details see the logfile /gpfs/alpine/mat226/proj-shared/Rishabh_copy/votca/csg-tutorials/spce/ibi_lammps/inverse.log #
#                                                                                                                          #
############################################################################################################################

Alternatively you can set the cg.inverse.lammps.command in your xml file.

I hope setting the path to my lammps executable is what is meant by this

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e4rUc%3Ds%2BBY1mYkGo28L9OwDJ0Fr7iBH5Hu05Hx8eUOaEw%40mail.gmail.com.

[Christoph Junghans]

On Thu, Jul 8, 2021 at 8:50 PM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

> I appreciate all the help but there is still a small roadblock I am facing.
>
> The xml file as you know has this setting by default :-
> <!-- general options for inverse script -->
> <inverse>
> <!-- votca units 300*0.00831451 kJ/mol -->
> <kBT>2.49435</kBT>
> <!-- use lammps as simulation program -->
> <program> lammps </program>
> <!-- lammps specific options -->
> <lammps>
> <!-- lammps script to run !-->
> <script>spce.in</script>
> <!-- topology to be used by csg_stat !-->
> <topol>spce.data</topol>
> <!-- traj file created by lammps !-->
> <traj>traj.dump</traj>

Sorry, I was a bit unclear, just add
<command> /gpfs/alpine/mat222/proj-shared/rdguha/lammps/build/lmp </command>
right here.

Alternatively, you could also add
/gpfs/alpine/mat222/proj-shared/rdguha/lammps/build to your PATH.

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktBmizoMbLLR3XLpse7cGE_q1GSOquqFHA2HDujSAB0xqQ%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

Thanks for the help with the lammps executable. VOTCA runs smoothly now for IBI. We were also looking into RE as we wanted an LJ analytical form for our CG potential. I am trying to modify the settings.xml script for re_lammps. For lj126, I changed the script as:-

 <re>
      <!-- function form for the potential (lj126 or ljg or cbspl)-->
      <function>lj126</function>
      <lj126>
        <!-- no. of knots for cbspl function -->
        <!-- nknots -->48<!-- /nknots -->
      </lj126>
    </re>

How should I format my CG-CG.param.init file now for lj126. I deleted the knot values and I just put in 2 best guess numbers for C12 and C6 in that file(PFA). But that is throwing up a format error:-

Generating potential tables from the initial parameters
Running critical command 'csg_reupdate --gentable true --param-in-ext param.new --options /gpfs/alpine/mat226/proj-shared/Rishabh_copy/votca/csg-tutorials/spce/re_lammps_lys_met/settings.xml'
an error occurred:
invaid type: file CG-CG.param.new, line 1

Callstack:
/sw/andes/spack-envs/base/opt/linux-rhel8-x86_64/gcc-9.3.0/votca-csg-2021.1-gkkujagr5jpheolnzjb4t3t2e43lxwbu/share/votca/scripts/inverse/inverse.sh - linenumber 156
    do_external - linenumber 177 in /sw/andes/spack-envs/base/opt/linux-rhel8-x86_64/gcc-9.3.0/votca-csg-2021.1-gkkujagr5jpheolnzjb4t3t2e43lxwbu/share/votca/scripts/inverse/functions_common.sh
        /sw/andes/spack-envs/base/opt/linux-rhel8-x86_64/gcc-9.3.0/votca-csg-2021.1-gkkujagr5jpheolnzjb4t3t2e43lxwbu/share/votca/scripts/inverse/prepare_re.sh - linenumber 39
            critical - linenumber 4 (see 'csg_call --cat function critical')
                die - linenumber 2 (see 'csg_call --cat function die')
###########################################################################################################################################################################################

#                                                                                                                                                                                         #
# ERROR:                                                                                                                                                                                  #

# critical: 'csg_reupdate --gentable true --param-in-ext param.new --options /gpfs/alpine/mat226/proj-shared/Rishabh_copy/votca/csg-tutorials/spce/re_lammps_lys_met/settings.xml' failed #
# For details see the logfile /gpfs/alpine/mat226/proj-shared/Rishabh_copy/votca/csg-tutorials/spce/re_lammps_lys_met/inverse.log                                                         #
#                                                                                                                                                                                         #
###########################################################################################################################################################################################

How should I format the param.init file? 

Thanks

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e5T7jNC7Ntsfdf4y3KYz5-su6cKPrsi8u7R1E3u5a2bGw%40mail.gmail.com.

[Christoph Junghans]

On Tue, Jul 13, 2021 at 9:13 AM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

The LJ126 inputs are not super-well documented and there is no
tutorial, but if I remember correctly, the param.init file needs to
look something like this:
1.0 <C6_value> i
2 <C12_value> i

I believe, the x values don't really matter, however I am not the
expert for this method, so maybe Sikandar can comment on this, too!

Christoph

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktBKMvOSzp-jaQ_q0Ewh0NdGS3K8baxx2RkA23Ga-QRL4g%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

Thanks for the update. Just FYI for future users, the x values do matter and the correct format is

0 <C6_value>  i

1 <C12_value> i

Can you or Sikander tell me what is the value of C6 or C12 here? Is it something like a best guess?

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e5F350Jm8_JDZRhphAGz1EbsunP4Q9Ai_mf3-8R62Bthg%40mail.gmail.com.

[Christoph Junghans]

On Tue, Jul 13, 2021 at 2:07 PM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

> Thanks for the update. Just FYI for future users, the x values do matter and the correct format is
> 0 <C6_value> i
> 1 <C12_value> i

Thanks! I believe these are best guesses.

If you get it working, please contribute an example to csg-tutorials
and/or write a bit of documentation (see csg/share/doc folder) for the
website.

Thanks,

Christoph

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktATu1ptirzBhrhX9ZsHr%2B7FkiizFYf%3DTd5gzQEZEWu8LQ%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph, 

I am currently working on it. I am struggling for it to match up with a target rdf I want it to match up to. A quick question:-

<re>
      <!-- function form for the potential (lj126 or ljg or cbspl)-->
      <function>lj126</function>
      <lj126>
        <!-- no. of knots for cbspl function -->
        <!-- nknots -->48<!-- /nknots -->
      </lj126>
    </re>

    <inverse>

In the cubic spline case we had to mention the number of knots in the settings file. Do we have to mention anything in the lj126 case? Or does the section just become:-

  <re>
      <!-- function form for the potential (lj126 or ljg or cbspl)-->
      <function>lj126</function>
      <lj126>

      </lj126>
    </re>  

Thanks

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e7zqoR2gzgzZxTkH1GQ%3D3k0U6-%2BoUr4Wor4G%3D_76w9ccg%40mail.gmail.com.

[Christoph Junghans]

On Tue, Jul 13, 2021 at 2:27 PM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

> I am currently working on it. I am struggling for it to match up with a target rdf I want it to match up to. A quick question:-
> <re>
> <!-- function form for the potential (lj126 or ljg or cbspl)-->
> <function>lj126</function>
> <lj126>
> <!-- no. of knots for cbspl function -->
> <!-- nknots -->48<!-- /nknots -->
> </lj126>
> </re>
> <inverse>
>
> In the cubic spline case we had to mention the number of knots in the settings file. Do we have to mention anything in the lj126 case? Or does the section just become:-
>
> <re>
> <!-- function form for the potential (lj126 or ljg or cbspl)-->
> <function>lj126</function>
> <lj126>
> </lj126>
> </re>

Looking at https://github.com/votca/csg/blob/master/src/tools/csg_reupdate.cc#L664_L670,
I don't think lj126 takes any options, so you can delete the whole
block.

Christoph

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktDkNe0wt6%3D1pfe8twu6tu5WPCQWZamXVTqRD42GPfJDZw%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

Sorry for another email. We are trying to get RE with LJ to work but the potentials are not converging. One question which we have is that we have to change the pair style from table linear to lj/cut. 

In this case, the pair coefficients won't be extracted from a table. Do you know of a way which can get us to change the values of epsilon , sigma and cutoffs in subsequent interations?

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e7q20Y78TKTrz85xw7gBVBX05pLdPNTWbMxiwpP29yYiw%40mail.gmail.com.

[Christoph Junghans]

.sh

On Tue, Jul 13, 2021 at 4:31 PM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

> Sorry for another email. We are trying to get RE with LJ to work but the potentials are not converging. One question which we have is that we have to change the pair style from table linear to lj/cut.
>
> In this case, the pair coefficients won't be extracted from a table. Do you know of a way which can get us to change the values of epsilon , sigma and cutoffs in subsequent interations?

Yes, the values should be in the param.cur file in each step directory.
You will have to customize the script to run the simulation though to
copy them out of the parameter file into the lammps input script.

To do this do the following:
1.) set <scriptpath> in your xml settings file (see propane/imc
tutorial), you can use $PWD, which means the same directory as
settings xml is in.
2.) create a file called csg_table in said <scriptpath> directory
3.) add a line to csg_table:
run lammps my_run_genericsim.sh
4.) copy run_genericsim.sh from votca/csg/share/script/inverse into
the <scriptpath> directory and rename it to my_run_genericsim
5.) modify my_run_genericsim.sh

For 5.) you want to add something like
sigma=$(awk '(NR==1){print $2}' param.cur)
eps=$(awk '(NR==2){print $2}' param.cur)
sed -e "s/@SIGMA@/${sigma}/" -e "s/@EPS@/${eps}/' -i lammps.script
somewhere before lammps is called.
This will extract sigma and eps from param.cur and then replace
@SIGMA@ and @EPS@ with the actual values in the simulation script.

Maybe lammps can read the coefficients from a file, too, but I am not sure here.

Christoph

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktBUHP5ekDHJ3Qai%3DrNNyaj6%2BauijAiSs%2BV%2BcqA1R9A9WA%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

Thanks for these pointers! They were helpful. We used a slightly different approach but we got the parameters to get updated. Can you clarify something for us?

We had previously discussed the parameters for the init file in case of lj126:-

The LJ126 inputs are not super-well documented and there is no
tutorial, but if I remember correctly, the param.init file needs to
look something like this:
1.0 <C6_value>  i
2 <C12_value> i

Can you confirm whether C6 and C12 are supposed to be what we assume it to be? i.e - C12 = 4*epsilon*sigma^12 and C6 = 4*epsilon*sigma^6? Or is c12 and C6 just sigma and epsilon here? I tried setting the values as C12 = 4*epsilon*sigma^12 and C6 = 4*epsilon*sigma^6. Firstly they are huge and secondly they barely change in subsequent iterations

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e6QMWzUZh%2BcqsdxkD3xXX8aX7%2BDBFvT24q6vbJoh7H%3DJQ%40mail.gmail.com.

[Christoph Junghans]

On Wed, Jul 14, 2021 at 2:10 PM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> Hey Christoph,
>

> Thanks for these pointers! They were helpful. We used a slightly different approach but we got the parameters to get updated. Can you clarify something for us?
>
> We had previously discussed the parameters for the init file in case of lj126:-
> The LJ126 inputs are not super-well documented and there is no
> tutorial, but if I remember correctly, the param.init file needs to
> look something like this:
> 1.0 <C6_value> i
> 2 <C12_value> i
>
> Can you confirm whether C6 and C12 are supposed to be what we assume it to be? i.e - C12 = 4*epsilon*sigma^12 and C6 = 4*epsilon*sigma^6? Or is c12 and C6 just sigma and epsilon here? I tried setting the values as C12 = 4*epsilon*sigma^12 and C6 = 4*epsilon*sigma^6. Firstly they are huge and secondly they barely change in subsequent iterations

Yes, it is all coefficients combined, see:
https://github.com/votca/csg/blob/master/src/libcsg/potentialfunctions/potentialfunctionlj126.cc#L29
lam_(0) and lam_(1) are the first and second parameters read from the file!

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktDC75cbx-oE8c2%2BDCWNfu%3D2bJAjMzepoU3_Vs4bVdBR%3Dw%40mail.gmail.com.

[Rishabh Guha]

Thanks! Is there anything which can basically increase the update rate / gradient per iteration?

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e4Pw0DgxSF5fBSn_Vts--LrG-mdgbxjNrRP5b0YgGp52g%40mail.gmail.com.

[Christoph Junghans]

On Wed, Jul 14, 2021 at 14:35 Rishabh Guha <rdg...@ncsu.edu> wrote:

Thanks! Is there anything which can basically increase the update rate / gradient per iteration?

What is the update rate?

Christoph 

[Rishabh Guha]

By update rate I mean the variable which is set as 'weight' in settings.xml file.

<post_add_options>
        <!-- convergence check options -->
        <convergence>
          <!-- for RE we check change in potentials/parameters (new-cur) -->
          <what>pot</what>
          <weight>1.0</weight>
          <base>cur</base>
          <norm>2</norm>
        </convergence>
        <average>
          <what>param pot</what>
        </average>
      </post_add_options>

Another thing which we wanted to discuss with you is this :-

The initial range of r (min and cutoff) are in nm in the settings file. So when we are calculating the lj potential:-

C12/r^12 - C6/r^6, the r is in nm.

In this case, C12 and C6 should also be in the respective units. So if our initial assumption is suppose sigma = 5A(0.5 nm) and epsilon = 0.1 kj/mol, the initial param.init values should be:-

0 0.0000975625 i
1 0.00625 i

If these are the initial units, we start getting negative values for CG-CG.pot from the second iteration and get an error. Are we setting these initial values correctly?

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e5dgyOgWA0NzfbxCGOZ_AVVUz-CyB9fQGoRs-EbnTjfqw%40mail.gmail.com.

[Christoph Junghans]

On Wed, Jul 14, 2021 at 6:30 PM Rishabh Guha <rdg...@ncsu.edu> wrote:
>
> By update rate I mean the variable which is set as 'weight' in settings.xml file.
> <post_add_options>
> <!-- convergence check options -->
> <convergence>
> <!-- for RE we check change in potentials/parameters (new-cur) -->
> <what>pot</what>
> <weight>1.0</weight>
> <base>cur</base>
> <norm>2</norm>
> </convergence>
> <average>
> <what>param pot</what>
> </average>
> </post_add_options>
>
> Another thing which we wanted to discuss with you is this :-
> The initial range of r (min and cutoff) are in nm in the settings file. So when we are calculating the lj potential:-
>
> C12/r^12 - C6/r^6, the r is in nm.
>
> In this case, C12 and C6 should also be in the respective units. So if our initial assumption is suppose sigma = 5A(0.5 nm) and epsilon = 0.1 kj/mol, the initial param.init values should be:-
>
> 0 0.0000975625 i
> 1 0.00625 i
>
> If these are the initial units, we start getting negative values for CG-CG.pot from the second iteration and get an error. Are we setting these initial values correctly?

Sorry, the C6, C12 parameter need to be in nm and kj/mol, too
0.1/0.5**6 and 0.1/0.5**12 (maybe I forgot a factor 4 here) and you
have to convert them before putting them in the lammps script.

> To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktDmssU84_hLUb-7VjitBhomuLxXOMvrj07uBwDTS%2BKnkg%40mail.gmail.com.

[Rishabh Guha]

So the numbers I mentioned in the previous email were obtained in nm and kj/mol.

C12 =  (4*0.1)/0.5**12 = 0.0000975625

C6 = (4*0.1)/0.5**6 = 0.00625

We are then using these commands to convert it back to angstrom and kj/mol in LAMMPS :-

C_12=$(awk '(NR==1){print $2}' CG-CG.param.cur)
C_06=$(awk '(NR==2){print $2}' CG-CG.param.cur)

sigma=`awk -v C_06=$C_06 -v C_12=$C_12 'BEGIN {printf "%0.5f", ((C_12/C_06)**(1.0/6.0))*10}'`
eps=`awk -v C_06=$C_06 -v C_12=$C_12 'BEGIN {printf "%0.5f", (C_06*C_06/4.0/C_12)}'`

sed -e "s/@SIGMA@/${sigma}/" -e "s/@EPS@/${eps}/" -i parm.in

But with these lines of codes the sigma and epsilon in first iteration is what we want - 5 A and 0.1,but as soon as the 2nd iteration starts we get negative values for epsilon:-

-4.55793 5.10840

I feel there is some conversion factor which we are messing up somewhere

[Christoph Junghans]

On Wed, Jul 14, 2021 at 20:21 Rishabh Guha <rdg...@ncsu.edu> wrote:

So the numbers I mentioned in the previous email were obtained in nm and kj/mol.

C12 =  (4*0.1)/0.5**12 = 0.0000975625

C6 = (4*0.1)/0.5**6 = 0.00625

We are then using these commands to convert it back to angstrom and kj/mol in LAMMPS :-

C_12=$(awk '(NR==1){print $2}' CG-CG.param.cur)
C_06=$(awk '(NR==2){print $2}' CG-CG.param.cur)

sigma=`awk -v C_06=$C_06 -v C_12=$C_12 'BEGIN {printf "%0.5f", ((C_12/C_06)**(1.0/6.0))*10}'`
eps=`awk -v C_06=$C_06 -v C_12=$C_12 'BEGIN {printf "%0.5f", (C_06*C_06/4.0/C_12)}'`

sed -e "s/@SIGMA@/${sigma}/" -e "s/@EPS@/${eps}/" -i parm.in

But with these lines of codes the sigma and epsilon in first iteration is what we want - 5 A and 0.1,but as soon as the 2nd iteration starts we get negative values for epsilon:-

-4.55793 5.10840

I feel there is some conversion factor which we are messing up somewhere

Sorry I have no good idea on that.

Christoph 

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAFKQktB5LGJjgssbKEZGxrAcPLRhV5jkmK65Mw%2BPSru2PU-MqQ%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph,

We are facing something very tricky with IBI in LAMMPS. The molecule we are working with is alanine with a MW of 89.1. When we use IBI with this molecule, after the lammps part is over we get this error:-

Simulation with lammps
Make update for ibi
Calculating rdfs with csg_stat using 32 tasks
################################################################################################################################################################################################################################################

#                                                                                                                                                                                                                                              #
# ERROR:                                                                                                                                                                                                                                       #

# critical: 'csg_stat --nt 32 --options /gpfs/alpine/mat226/proj-shared/Rishabh_copy/lammps_files_created_rishabh/check_abc/03Prod/00_VOTCA_IBI/alanine_alanine/settings.xml --top spce.data --trj traj.dump --begin 0 --first-frame 0' failed #
# For details see the logfile /gpfs/alpine/mat226/proj-shared/Rishabh_copy/lammps_files_created_rishabh/check_abc/03Prod/00_VOTCA_IBI/alanine_alanine/inverse.log                                                                              #
#                                                                                                                                                                                                                                              #
################################################################################################################################################################################################################################################

When I investigated the log file, I found out that while reading the data file, VOTCA is not recognizing the bead:-

WARNING: The votca lammps data reader is only able to read lammps files formatted in the following styles:
angle
atom
bond
full
molecule

These styles use the following formats in the atom block:
atom-ID molecule-ID atom-type charge x y z
atom-ID molecule-ID atom-type charge x y z nx ny nz
atom-ID molecule-ID atom-type x y z
atom-ID molecule-ID atom-type x y z nx ny nz
atom-ID atom-type x y z
atom-ID atom-type x y z nx ny nz

I have 500 beads in 500 molecules

WARNING: The votca lammps data reader is only able to read lammps files formatted in the following styles:
angle
atom
bond
full
molecule

These styles use the following formats in the atom block:
atom-ID molecule-ID atom-type charge x y z
atom-ID molecule-ID atom-type charge x y z nx ny nz
atom-ID molecule-ID atom-type x y z
atom-ID molecule-ID atom-type x y z nx ny nz
atom-ID atom-type x y z
atom-ID atom-type x y z nx ny nz

But when we increase the weight to 99.1 instead of 89.1, the simulations run smoothly:-

Running critical command 'csg_stat --nt 32 --options /gpfs/alpine/mat226/proj-shared/Rishabh_copy/lammps_files_created_rishabh/check_abc/03Prod/00_VOTCA_IBI/alanine_alanine/settings.xml --top spce.data --trj traj.dump --begin 0 --first-frame 0'
begin to calculate distribution functions
# of bonded interactions: 0
# of non-bonded interactions: 1

WARNING: The votca lammps data reader is only able to read lammps files formatted in the following styles:
angle
atom
bond
full
molecule

These styles use the following formats in the atom block:
atom-ID molecule-ID atom-type charge x y z
atom-ID molecule-ID atom-type charge x y z nx ny nz
atom-ID molecule-ID atom-type x y z
atom-ID molecule-ID atom-type x y z nx ny nz
atom-ID atom-type x y z
atom-ID atom-type x y z nx ny nz

Unable to associate mass 99.1 with element assuming pseudo atom, assigning name Bead1 .

I have 500 beads in 500 molecules

Do you know why certain weights are getting recognized by VOTCA while some aren't?

Regards

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e6a3eUVYGDLOhbZejo_NqPt32r7uBmdjDB2JAkjXVksVg%40mail.gmail.com.

[Christoph Junghans]

On Tue, Aug 3, 2021 at 16:51 Rishabh Guha <rdg...@ncsu.edu> wrote:

Hey Christoph,

the above are just warnings, can you run the csg_stat command from above by hand in the last step directory and look for the error message.

The command should be:

csg_stat --nt 32 --options /gpfs/alpine/mat226/proj-shared/Rishabh_copy/lammps_files_created_rishabh/check_abc/03Prod/00_VOTCA_IBI/alanine_alanine/settings.xml --top spce.data --trj traj.dump --begin 0 --first-frame 0

Regards

[Rishabh Guha]

Hey Christoph,

Thanks for the reply,

So as we were experiencing yesterday, when I set the mass as 89.1 (real mass of alanine), this is the error we get (I ran csg_stat manually to get this):

an error occurred:
Topology does not have beads of type "Bead1"
This was specified in type1 of interaction "CG1-CG1"

When I change the mass to 99.1 csg-stat runs smoothly:-

Reading frame, timestep 10000
Reading frame, timestep 10100
Reading frame, timestep 10200
Reading frame, timestep 10300
Reading frame, timestep 10400
Reading frame, timestep 10500
Reading frame, timestep 10600
Reading frame, timestep 10700
Reading frame, timestep 10800
Reading frame, timestep 10900
Reading frame, timestep 11000
Reading frame, timestep 11100
Reading frame, timestep 11200
Reading frame, timestep 11300
Reading frame, timestep 11400
Reading frame, timestep 11500
Reading frame, timestep 11600
Reading frame, timestep 11700
Reading frame, timestep 11800
Reading frame, timestep 11900

.....

It seems like when VOTCA is encountering certain numbers, it's failing to read them

To view this discussion on the web visit https://groups.google.com/d/msgid/votca/CAHG27e6E5rZioLPmPpd%2BOhfopOwAv9Ajmy6EXi11upzPraZNOQ%40mail.gmail.com.

[Rishabh Guha]

Hey Christoph, 

Just as a follow up, there is something surely off with the data reading capabilities of csg-stat when it comes to lammps atleast: it is not specific to our systems. If you go to the csg-tutorials/spce/ibi_lammps folder for votca and try to run the ibi_lammps with a hypothetical weight of 88.15 instead of 18.15, you will get the same error. For some reason, the reader is reading the mass if they are some specific numbers while not reading them if it's some different number.

Regards

[Christoph Junghans]

Ok, that sounds like a real issue, can you make a minimal reproducible example and open an issue at: 

https://github.com/votca/csg/issues

Please attach all input files and details on the command to run.

Instead of the full trajectory you should be able to use a trajectory with just one frame.

We will have a look soon.

Christoph