normal Reverse CG with Amber

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 2 months ago #475 by is
Reverse CG with Amber was created by is
Hi,

I have performed an all-atom simulation of a small system containing two-domain protein, water molecules and Na+ ions, using Amber03 ff. Based on the output structure I did CG simulation. Now I would like to make a reverse transformation of this CG system, however as I used Amber I have problems with reversing CG water beads into FG water molecules.

Using:

g_fg2cg -pfg aa.top -pcg cg.top -n 0 -c last.gro -o fg.gro


I get an error:

Fatal error:
Atomtype 'OW' not found!

How can I change fg_w.itp file to match Amber nomenclature?

Thank you for your help
is

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #476 by xavier
Replied by xavier on topic Reverse CG with Amber

is wrote: I have performed an all-atom simulation of a small system containing two-domain protein, water molecules and Na+ ions, using Amber03 ff. Based on the output structure I did CG simulation. Now I would like to make a reverse transformation of this CG system, however as I used Amber I have problems with reversing CG water beads into FG water molecules.

Using:

g_fg2cg -pfg aa.top -pcg cg.top -n 0 -c last.gro -o fg.gro


I get an error:

Fatal error:
Atomtype 'OW' not found!

How can I change fg_w.itp file to match Amber nomenclature?

What type of water did you use? Can't you just change it to
the pone that would be fine?

Did you manage to implement the mapping for the Amber force field?



Thank you for your help
is

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #477 by is
Replied by is on topic Reverse CG with Amber

xavier wrote:

is wrote: I have performed an all-atom simulation of a small system containing two-domain protein, water molecules and Na+ ions, using Amber03 ff. Based on the output structure I did CG simulation. Now I would like to make a reverse transformation of this CG system, however as I used Amber I have problems with reversing CG water beads into FG water molecules.

Using:

g_fg2cg -pfg aa.top -pcg cg.top -n 0 -c last.gro -o fg.gro


I get an error:

Fatal error:
Atomtype 'OW' not found!

How can I change fg_w.itp file to match Amber nomenclature?

What type of water did you use? Can't you just change it to
the pone that would be fine?


Did you manage to implement the mapping for the Amber force field?



Thank you for your help
is


I used SPC water model.
What do you mean by "the pone"?
No I did not manage to implement mapping for the AMBER ff, I have only installed it in that modified Gromacs version. If you have any suggestions about mapping for the AMBER ff, I would really appreciate them.

Best
is

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #478 by is
Replied by is on topic Reverse CG with Amber
So I managed to use Amber for mapping water molecules, but now I have another problem. I get a segmentation fault about wrong naming of water molecules. However, I checked cg .top file, as suggested, also other files used to convert cg to fg and I can't find what is wrong. Naming seems to be correct. What else can cause this segmentation fault?

Generated 2701 of the 2701 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 0.5
Generated 2701 of the 2701 1-4 parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
Excluding 3 bonded neighbours for Protein_B 1
Excluding 0 bonded neighbours for FG_W 8736
Excluding 1 bonded neighbours for NA+ 9
# BONDS: 5221
# HARMONIC: 34944
# ANGLES: 9456
# PDIHS: 1014
# RBDIHS: 10726
# LJ14: 13592
# CONSTR: 104832
calling cpp...
processing topology...
Generated 4 of the 780 non-bonded parameter combinations
Excluding 1 bonded neighbours for Protein_A 1
Excluding 1 bonded neighbours for Protein_B 1
Excluding 1 bonded neighbours for W 8736
Excluding 1 bonded neighbours for NA+ 9

Number of fg atoms 110014
Number of cg atoms 9458
Reading frames from gro file 'Generated by trjconv : Protein in water t= 998400.00000', 9458 atoms.
Reading frame 0 time 998400.000 1295956981
Name of CG water in topology should be W
Segmentation fault

Thank you
is

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #479 by xavier
Replied by xavier on topic Reverse CG with Amber

I used SPC water model.
What do you mean by "the pone"?

I meant "the one" ...

No I did not manage to implement mapping for the AMBER ff, I have only installed it in that modified Gromacs version.

I am not sure what is the difference between implement and install?
You got to choose a mapping and implement/install it in the version of gromacs that runs the
reverse transformation ...

If you have any suggestions about mapping for the AMBER ff, I would really appreciate them.

Best
is

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #480 by xavier
Replied by xavier on topic Reverse CG with Amber

is wrote: So I managed to use Amber for mapping water molecules, but now I have another problem. I get a segmentation fault about wrong naming of water molecules. However, I checked cg .top file, as suggested, also other files used to convert cg to fg and I can't find what is wrong. Naming seems to be correct. What else can cause this segmentation fault?

Could it be that the atomistic and cg topologies are equivalent and the code expects a CG water but something else is on the way?

Generated 2701 of the 2701 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 0.5
Generated 2701 of the 2701 1-4 parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
Excluding 3 bonded neighbours for Protein_B 1
Excluding 0 bonded neighbours for FG_W 8736
Excluding 1 bonded neighbours for NA+ 9
# BONDS: 5221
# HARMONIC: 34944
# ANGLES: 9456
# PDIHS: 1014
# RBDIHS: 10726
# LJ14: 13592
# CONSTR: 104832
calling cpp...
processing topology...
Generated 4 of the 780 non-bonded parameter combinations
Excluding 1 bonded neighbours for Protein_A 1
Excluding 1 bonded neighbours for Protein_B 1
Excluding 1 bonded neighbours for W 8736
Excluding 1 bonded neighbours for NA+ 9

Number of fg atoms 110014
Number of cg atoms 9458
Reading frames from gro file 'Generated by trjconv : Protein in water t= 998400.00000', 9458 atoms.
Reading frame 0 time 998400.000 1295956981
Name of CG water in topology should be W
Segmentation fault

Thank you
is

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #481 by is
Replied by is on topic Reverse CG with Amber
molecules in topologies files are equivalent
atomistic:

[ molecules ]
; name number
Protein_A 1
Protein_B 1
FG_W 8736
NA+ 9

cg:
[ molecules ]
; name number
Protein_A 1
Protein_B 1
W 8736
NA+ 9

of course they include different .itp files

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #482 by xavier
Replied by xavier on topic Reverse CG with Amber

is wrote: molecules in topologies files are equivalent
atomistic:

[ molecules ]
; name number
Protein_A 1
Protein_B 1
FG_W 8736
NA+ 9

cg:
[ molecules ]
; name number
Protein_A 1
Protein_B 1
W 8736
NA+ 9

of course they include different .itp files


I would imagine that you have to define 4 times less CG water bead than in an atomistic system!

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #483 by is
Replied by is on topic Reverse CG with Amber
In cg top file I have 8736 water beads, while in atomistic 31554 water molecules (before changing anything). That is why I replaced number of water molecules in atomistic simulation by number of water beads.
I changed it and it works.
Thank you for your help.

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #484 by andrzej
Replied by andrzej on topic Reverse CG with Amber
The number of FG_W in aa.top and W cg.top should be the same, but you should also have mapping for ions.

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #485 by is
Replied by is on topic Reverse CG with Amber
Hi,

Converting from cg to fg using g_fg2cg works, however I get the similar error as described in this topic: md.chem.rug.nl/cgmartini/index.php/user-...rm/forum/topic?id=19 .
I get also zero coordinates for the protein, some part of water molecules and all Na+ ions:

1NASP N 1 0.000 0.000 0.000
1NASP H1 2 0.000 0.000 0.000
1NASP H2 3 0.000 0.000 0.000
1NASP H3 4 0.000 0.000 0.000
1NASP CA 5 0.000 0.000 0.000
1NASP HA 6 0.000 0.000 0.000
1NASP CB 7 0.000 0.000 0.000
1NASP HB1 8 0.000 0.000 0.000
1NASP HB2 9 0.000 0.000 0.000
1NASP CG 10 0.000 0.000 0.000
1NASP OD1 11 0.000 0.000 0.000
1NASP OD2 12 0.000 0.000 0.000

I followed advice from quoted topic, and moving ions in cg .gro file before water didn't work. Also, when I replaced number of water molecules in fg.top by number of cg water beads, to make them equal(as suggested by Andrzej in that topic) I get the same error I posted in the beginning (segmentation fault). I got confused due to this contradictory information, it seems that when this numbers are different I can avoid segmentation fault/ however there might be something else causing this error.
Have you any suggestions what can it be?
What can I do to handle problem with zero coordinates?

Best
is

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #486 by is
Replied by is on topic Reverse CG with Amber

andrzej wrote: The number of FG_W in aa.top and W cg.top should be the same, but you should also have mapping for ions.


I see we posted in the same time.
I have only 9 Na+ ions used for neutralizing the system, and 9 beads in cg .top file. I saw that you asked a question (in one of the topics) about mapping section in ion itp file and how to construct it. Did you received an answer to these questions?

Best
is

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #487 by andrzej
Replied by andrzej on topic Reverse CG with Amber
Hey

When You change na+ to be a water for the time of reconstruction that should do the trick. So the topology should look like

[ molecules ]
; name number
Protein_A 1
Protein_B 1
FG_W 8745


cg:
[ molecules ]
; name number
Protein_A 1
Protein_B 1
W 8745


Then you can put back this 9 ions after reconstruction. CG ion is 6 aa waters and the ion, so this should be the mapping, however there is missing function for the keeping this 7 molecules together in one bead. I will make the version with additional function available next week if you prefer this option.

Andrzej

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #488 by is
Replied by is on topic Reverse CG with Amber

andrzej wrote: Hey

When You change na+ to be a water for the time of reconstruction that should do the trick. So the topology should look like

[ molecules ]
; name number
Protein_A 1
Protein_B 1
FG_W 8745


cg:
[ molecules ]
; name number
Protein_A 1
Protein_B 1
W 8745


Then you can put back this 9 ions after reconstruction. CG ion is 6 aa waters and the ion, so this should be the mapping, however there is missing function for the keeping this 7 molecules together in one bead. I will make the version with additional function available next week if you prefer this option.

Andrzej


So I replaced Na+ by W, but still I get a segmentation fault:

Generated 2701 of the 2701 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 0.5
Generated 2701 of the 2701 1-4 parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
Excluding 3 bonded neighbours for Protein_B 1
Excluding 0 bonded neighbours for FG_W 8745
NOTE:
System has non-zero total charge: -9.000003e+00

# BONDS: 5221
# HARMONIC: 34980
# ANGLES: 9456
# PDIHS: 1014
# RBDIHS: 10726
# LJ14: 13592
# CONSTR: 104940
calling cpp...
processing topology...
Generated 4 of the 780 non-bonded parameter combinations
Excluding 1 bonded neighbours for Protein_A 1
Excluding 1 bonded neighbours for Protein_B 1
Excluding 1 bonded neighbours for W 8745
NOTE:
System has non-zero total charge: -9.000000e+00


Number of fg atoms 110113
Number of cg atoms 9458
Reading frames from gro file 'Generated by trjconv : Protein in water t= 998400.00000', 9458 atoms.
Reading frame 0 time 998400.000 1296054071
Name of CG water in topology should be W
Segmentation fault

What is odd is that the fg.gro file is generated after I increase number of water molecules in fg.top file to 9458 (or more) which is total number of beads in the system.

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #489 by xavier
Replied by xavier on topic Reverse CG with Amber

is wrote:

andrzej wrote: Hey

When You change na+ to be a water for the time of reconstruction that should do the trick. So the topology should look like

[ molecules ]
; name number
Protein_A 1
Protein_B 1
FG_W 8745


cg:
[ molecules ]
; name number
Protein_A 1
Protein_B 1
W 8745


Then you can put back this 9 ions after reconstruction. CG ion is 6 aa waters and the ion, so this should be the mapping, however there is missing function for the keeping this 7 molecules together in one bead. I will make the version with additional function available next week if you prefer this option.

Andrzej


So I replaced Na+ by W, but still I get a segmentation fault:

Generated 2701 of the 2701 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 0.5
Generated 2701 of the 2701 1-4 parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
Excluding 3 bonded neighbours for Protein_B 1
Excluding 0 bonded neighbours for FG_W 8745
NOTE:
System has non-zero total charge: -9.000003e+00

# BONDS: 5221
# HARMONIC: 34980
# ANGLES: 9456
# PDIHS: 1014
# RBDIHS: 10726
# LJ14: 13592
# CONSTR: 104940
calling cpp...
processing topology...
Generated 4 of the 780 non-bonded parameter combinations
Excluding 1 bonded neighbours for Protein_A 1
Excluding 1 bonded neighbours for Protein_B 1
Excluding 1 bonded neighbours for W 8745
NOTE:
System has non-zero total charge: -9.000000e+00


Number of fg atoms 110113
Number of cg atoms 9458
Reading frames from gro file 'Generated by trjconv : Protein in water t= 998400.00000', 9458 atoms.
Reading frame 0 time 998400.000 1296054071
Name of CG water in topology should be W
Segmentation fault

What is odd is that the fg.gro file is generated after I increase number of water molecules in fg.top file to 9458 (or more) which is total number of beads in the system.


What happens if you put the correct number of FG waters?

Andrzej: why should it be the same number? It is probably very confusing for people!

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #490 by andrzej
Replied by andrzej on topic Reverse CG with Amber
There still must be something wrong in your topology. You can debug it by parts, so first check if g_cg2fg works only for the pep_A, then for pep_A + pep_B , at the end add water. When the mappings are correctly set and there is no ions, everything should run smoothly.

In the fg_w.itp four atomistic water molecules are bundled into one artificial water "cluster" which corresponds to one CG water. Thus the number of water molecules in topology notation is the same.

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #491 by is
Replied by is on topic Reverse CG with Amber

xavier wrote:

is wrote:

andrzej wrote: Hey

When You change na+ to be a water for the time of reconstruction that should do the trick. So the topology should look like

[ molecules ]
; name number
Protein_A 1
Protein_B 1
FG_W 8745


cg:
[ molecules ]
; name number
Protein_A 1
Protein_B 1
W 8745


Then you can put back this 9 ions after reconstruction. CG ion is 6 aa waters and the ion, so this should be the mapping, however there is missing function for the keeping this 7 molecules together in one bead. I will make the version with additional function available next week if you prefer this option.

Andrzej


So I replaced Na+ by W, but still I get a segmentation fault:

Generated 2701 of the 2701 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 0.5
Generated 2701 of the 2701 1-4 parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
Excluding 3 bonded neighbours for Protein_B 1
Excluding 0 bonded neighbours for FG_W 8745
NOTE:
System has non-zero total charge: -9.000003e+00

# BONDS: 5221
# HARMONIC: 34980
# ANGLES: 9456
# PDIHS: 1014
# RBDIHS: 10726
# LJ14: 13592
# CONSTR: 104940
calling cpp...
processing topology...
Generated 4 of the 780 non-bonded parameter combinations
Excluding 1 bonded neighbours for Protein_A 1
Excluding 1 bonded neighbours for Protein_B 1
Excluding 1 bonded neighbours for W 8745
NOTE:
System has non-zero total charge: -9.000000e+00


Number of fg atoms 110113
Number of cg atoms 9458
Reading frames from gro file 'Generated by trjconv : Protein in water t= 998400.00000', 9458 atoms.
Reading frame 0 time 998400.000 1296054071
Name of CG water in topology should be W
Segmentation fault

What is odd is that the fg.gro file is generated after I increase number of water molecules in fg.top file to 9458 (or more) which is total number of beads in the system.


What happens if you put the correct number of FG waters?

Andrzej: why should it be the same number? It is probably very confusing for people!


So, if I put the same num. of FG_W and W that is 8736 or 8745 (when I replace Na+ by W), in both cases I get:

Generated 2701 of the 2701 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 0.5
Generated 2701 of the 2701 1-4 parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
Excluding 3 bonded neighbours for Protein_B 1
Excluding 0 bonded neighbours for FG_W 8745
NOTE:
System has non-zero total charge: -9.000003e+00

# BONDS: 5221
# HARMONIC: 34980
# ANGLES: 9456
# PDIHS: 1014
# RBDIHS: 10726
# LJ14: 13592
# CONSTR: 104940
calling cpp...
processing topology...
Generated 4 of the 780 non-bonded parameter combinations
Excluding 1 bonded neighbours for Protein_A 1
Excluding 1 bonded neighbours for Protein_B 1
Excluding 1 bonded neighbours for W 8745
NOTE:
System has non-zero total charge: -9.000000e+00


Number of fg atoms 110113
Number of cg atoms 9458
Reading frames from gro file 'Generated by trjconv : Protein in water t= 998400.00000', 9458 atoms.
Reading frame 0 time 998400.000 1296056504
Name of CG water in topology should be W
Segmentation fault

If I assign 8736*4 for FG_W and 8736 for W, fg.gro file is generated but the protein, Na+ and some of water molecules coordinates are zero.

Please Log in or Create an account to join the conversation.

More
8 years 1 month ago #492 by xavier
Replied by xavier on topic Reverse CG with Amber
Andrzej should know better than me. So you should follow his advices ...

I am not quite sure though how to replace the NA+ by water ... Andrzej may be more specific about this?

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #493 by is
Replied by is on topic Reverse CG with Amber

andrzej wrote: There still must be something wrong in your topology. You can debug it by parts, so first check if g_cg2fg works only for the pep_A, then for pep_A + pep_B , at the end add water. When the mappings are correctly set and there is no ions, everything should run smoothly.

In the fg_w.itp four atomistic water molecules are bundled into one artificial water "cluster" which corresponds to one CG water. Thus the number of water molecules in topology notation is the same.


I included only Protein_A in both top files and gro file, and I get segmentation fault:
Generated 2701 of the 2701 non-bonded parameter combinations
Generating 1-4 interactions: fudge = 0.5
Generated 2701 of the 2701 1-4 parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
NOTE:
System has non-zero total charge: -9.000003e+00

# BONDS: 5200
# ANGLES: 9417
# PDIHS: 1013
# RBDIHS: 10681
# LJ14: 13541
calling cpp...
processing topology...
Generated 4 of the 780 non-bonded parameter combinations
Excluding 1 bonded neighbours for Protein_A 1
NOTE:
System has non-zero total charge: -9.000000e+00


Number of fg atoms 5151
Number of cg atoms 711
Reading frames from gro file 'Giving Russians Opium May Alter Current Situation', 711 atoms.
Reading frame 0 time 0.000 1296061305
Name of CG water in topology should be W
Segmentation fault

Those are my top files:
fg:

; Include forcefield parameters
#include "ffamber03.itp"

; Include chain topologies
#include "prot_A.itp"

[ system ]
; Name
Protein

[ molecules ]
; Compound #mols
Protein_A 1

and cg:

; Include forcefield parameters
#include "martini_v2.1.itp"
#include "prot_cg.itp"

[ system ]
; name
Protein

[ molecules ]
; name number
Protein_A 1

So they, as well as cg .gro file, look fine. What can be wrong?

is

Please Log in or Create an account to join the conversation.

  • is
  • is's Avatar Topic Author
  • Offline
  • Senior Boarder
More
8 years 1 month ago #494 by is
Replied by is on topic Reverse CG with Amber

xavier wrote: Andrzej should know better than me. So you should follow his advices ...

I am not quite sure though how to replace the NA+ by water ... Andrzej may be more specific about this?


Thank you for your time and suggestions.

is

Please Log in or Create an account to join the conversation.

Time to create page: 0.125 seconds