- Posts: 36
Reverse transformation
- nivedita
-
Topic Author
- Offline
- Junior Boarder
Less
More
11 years 4 weeks ago #1501
by nivedita
Reverse transformation was created by nivedita
Hello All,
I did cg simulation in gromacs 4.5.4.Now for doing reverse transform i downloaded and install special gromacs version 3.3.1 having g_cg2fg file but im unable to do reverse transform.
i extract the trajectory(from cg simulation)named last_frame.pdb, i used this file for mapping by using command
pdb2gmx -f last_frame.pdb -missing then i select GROMOS96 43a2 FF but it shows error Incomplete ring in HIS95 again and again.
will you please tell which structure i have to mapped? the original all_atom pdb structure or structure which i get after cg simulation?
I did cg simulation in gromacs 4.5.4.Now for doing reverse transform i downloaded and install special gromacs version 3.3.1 having g_cg2fg file but im unable to do reverse transform.
i extract the trajectory(from cg simulation)named last_frame.pdb, i used this file for mapping by using command
pdb2gmx -f last_frame.pdb -missing then i select GROMOS96 43a2 FF but it shows error Incomplete ring in HIS95 again and again.
will you please tell which structure i have to mapped? the original all_atom pdb structure or structure which i get after cg simulation?
Please Log in or Create an account to join the conversation.
- bt12d026
-
- Visitor
11 years 3 weeks ago #1517
by bt12d026
Replied by bt12d026 on topic Reverse transformation
Hi,
I suspect you should use g_cg2fg tool. Please refer to reverse transformation in the tutorial section
I suspect you should use g_cg2fg tool. Please refer to reverse transformation in the tutorial section
Please Log in or Create an account to join the conversation.
- nivedita
-
Topic Author
- Offline
- Junior Boarder
Less
More
- Posts: 36
11 years 3 weeks ago #1518
by nivedita
Replied by nivedita on topic Reverse transformation
i already follow the tutorial and i mention my steps and error, if you see my earlier post but it not work, now i again trying to do that.my protein is simple not membranous.if anybody know how to do,and where i do modification in parameter please reply me as soon as possible.
Thanks in advance
Thanks in advance
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
11 years 3 weeks ago #1519
by Clement
Replied by Clement on topic Reverse transformation
Hi nivedita,
To perform the reverse transformation, you need to have the CG conformation you want to convert into atomistic (your last_frame.pdb), a CG topology (the one you created/used to perform your CG simulation) and an atomistic (AA) topology (be sure to create it with the pdb2gmx from the tweaked version of gromacs, some dihedral restraints need to be added). Then you use the g_cg2fg to create a starting AA structure from your CG one, and perform a simulated annealing on it to smoothly reach a reasonable AA conformation.
Good luck with that! :-)
P.S.: thanks Venkat!
To perform the reverse transformation, you need to have the CG conformation you want to convert into atomistic (your last_frame.pdb), a CG topology (the one you created/used to perform your CG simulation) and an atomistic (AA) topology (be sure to create it with the pdb2gmx from the tweaked version of gromacs, some dihedral restraints need to be added). Then you use the g_cg2fg to create a starting AA structure from your CG one, and perform a simulated annealing on it to smoothly reach a reasonable AA conformation.
Good luck with that! :-)
P.S.: thanks Venkat!
Please Log in or Create an account to join the conversation.
- nivedita
-
Topic Author
- Offline
- Junior Boarder
Less
More
- Posts: 36
11 years 3 weeks ago #1520
by nivedita
Replied by nivedita on topic Reverse transformation
hello clement,
when i run my first step pdb2gmx -f filename.pdb -missing
then it show that "Total charge 2.000 e" and generate conf.gro, topol.top and posre.itp be default.now what i have to do?Is that i have to run editconf to create a box then genbox to add water molecule then genion, grompp to neutralize that charge , if yes then what water model i have to use ?
if anything wrong then plz mention that also.
Thanks for your reply
when i run my first step pdb2gmx -f filename.pdb -missing
then it show that "Total charge 2.000 e" and generate conf.gro, topol.top and posre.itp be default.now what i have to do?Is that i have to run editconf to create a box then genbox to add water molecule then genion, grompp to neutralize that charge , if yes then what water model i have to use ?
if anything wrong then plz mention that also.
Thanks for your reply
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
11 years 3 weeks ago #1521
by Clement
Replied by Clement on topic Reverse transformation
You don't want to create/solvate/neutralize a new simulation box, you want to reverse your CG conformation to AA: the new box of simulation (the new coordinates) will be deducted from the CG coordinates. The only thing you need for the reverse transformation is the AA protein topology, which you just created with pdb2gmx.
The command line extracted from the tutorial is:
g_fg2cg -pfg fg.top -pcg cg.top -c cg.gro -o fg.gro -n 0
fg.top being the AA system topology, cg.top the CG system topology, cg.gro the CG conformation you want to reverse-transform, fg.gro the result of this transformation. The flag -n 0 is here to show you're doing a CG > FG transformation (-n 1 is for FG > CG).
When you'll write your AA system topology (including the protein topology you just created/number of water/ions/whatever is present in your system), don't forget to multiply the number of water by 4 (1 CG water = 4 AA water molecules), and add to that 4 times the number of CG ions (1 CG ion = 1 AA ion + 4 CG water molecules).
You're getting closer!
The command line extracted from the tutorial is:
g_fg2cg -pfg fg.top -pcg cg.top -c cg.gro -o fg.gro -n 0
fg.top being the AA system topology, cg.top the CG system topology, cg.gro the CG conformation you want to reverse-transform, fg.gro the result of this transformation. The flag -n 0 is here to show you're doing a CG > FG transformation (-n 1 is for FG > CG).
When you'll write your AA system topology (including the protein topology you just created/number of water/ions/whatever is present in your system), don't forget to multiply the number of water by 4 (1 CG water = 4 AA water molecules), and add to that 4 times the number of CG ions (1 CG ion = 1 AA ion + 4 CG water molecules).
You're getting closer!
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
11 years 3 weeks ago #1522
by Clement
Replied by Clement on topic Reverse transformation
Hang on, why did you use the "-missing" flag in your pdb2gmx? Don't use that! Fix your AA protein structure before anything else... Nothing will work otherwise. Did you create your CG topology from this same PDB? What file do you use to create the AA topology?
Please Log in or Create an account to join the conversation.
- nivedita
-
Topic Author
- Offline
- Junior Boarder
Less
More
- Posts: 36
11 years 3 weeks ago #1523
by nivedita
Replied by nivedita on topic Reverse transformation
hello clement,
when i simply use pdb2gmx -f filename.pdb and choose GROMOS96 43a2 force field then it will show Fatal error:
Atom H not found in residue 1 while adding mapping
but when i use pdb2gmx -f filename.pdb -missing then it will run without any error.
when i simply use pdb2gmx -f filename.pdb and choose GROMOS96 43a2 force field then it will show Fatal error:
Atom H not found in residue 1 while adding mapping
but when i use pdb2gmx -f filename.pdb -missing then it will run without any error.
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
11 years 3 weeks ago #1524
by Clement
Replied by Clement on topic Reverse transformation
Did you try to fix that? What kind of residue is it? Which hydrogen is missing?
This flag shouldn't be used lightly...
This flag shouldn't be used lightly...
Please Log in or Create an account to join the conversation.
- xavier
-
- Offline
- Admin
Less
More
- Posts: 416
11 years 3 weeks ago #1525
by xavier
Replied by xavier on topic Reverse transformation
This error is due to your atomistic pdb file. Try to use the -ignh flag. It will ignore the hydrogen persent in the pdb file and construct the ones it needs.
nivedita wrote: hello clement,
when i simply use pdb2gmx -f filename.pdb and choose GROMOS96 43a2 force field then it will show Fatal error:
Atom H not found in residue 1 while adding mapping
but when i use pdb2gmx -f filename.pdb -missing then it will run without any error.
Please Log in or Create an account to join the conversation.
- nivedita
-
Topic Author
- Offline
- Junior Boarder
Less
More
- Posts: 36
11 years 3 weeks ago #1526
by nivedita
Replied by nivedita on topic Reverse transformation
hello all,
when i use pdb2gmx -f filename.pdb -ignh then again i got the same error
Fatal error:
Atom H not found in residue 1 while adding mapping and the first residue is glutamine (GLN).one more thing is that i already did all-atom simulation using same molecule in gromacs 4.5.5 there is no problem in running pdb2gmx.
when i use pdb2gmx -f filename.pdb -ignh then again i got the same error
Fatal error:
Atom H not found in residue 1 while adding mapping and the first residue is glutamine (GLN).one more thing is that i already did all-atom simulation using same molecule in gromacs 4.5.5 there is no problem in running pdb2gmx.
Please Log in or Create an account to join the conversation.
- nivedita
-
Topic Author
- Offline
- Junior Boarder
Less
More
- Posts: 36
11 years 3 weeks ago #1527
by nivedita
Replied by nivedita on topic Reverse transformation
Hello clement,
i finally used pdb2gmx -f filename.pdb -missing by default it will generate topol.top ,conf.gro and posre.itp files then i did
g_fg2cg -pfg topol.top -pcg system.top -c last_frame.pdb -n 0 -o fg.gro -wat 1
it will generate fg.gro file having coordinate. now i modify fg.mdp file and run further step i.e.,grompp -f fg.mdp -c fg.gro -p topol.top -o topol.tpr then it was showing error:
creating statusfile for 1 node...
' for variable integrator, using 'md'
Next time use one of: 'md' 'steep' 'cg' 'bd' 'sd' 'nm' 'l-bfgs' 'tpi'
' for variable ns-type, using 'Grid'
Next time use one of: 'Grid' 'Simple'
' for variable coulombtype, using 'Cut-off'ield
Next time use one of: 'Cut-off' 'Reaction-Field' 'Generalized-Reaction-Field' 'PM E' 'Ewald' 'PPPM' 'Poisson' 'Switch' 'Shift' 'User' 'Generalized-Born' 'Reaction- Field-nec' 'Encad-shift' 'PME-User'
' for variable tcoupl, using 'No'
Next time use one of: 'No' 'Berendsen' 'Nose-Hoover' 'yes' 'Andersen' 'Andersen-i nterval'
' for variable Pcoupl, using 'No'
Next time use one of: 'No' 'Berendsen' 'Parrinello-Rahman' 'Isotropic'
' for variable gen-vel, using 'no'
Next time use one of: 'no' 'yes'
' for variable constraints, using 'none'
Next time use one of: 'none' 'h-bonds' 'all-bonds' 'h-angles' 'all-angles'
' for variable rel_water, using 'no'
Next time use one of: 'no' 'yes'
checking input for internal consistency...
...ling /lib/cpp
: No such file or directory
cpp exit code: 32512
-I/usr/local/gromacs/share/gromacs/top topol.top > gromppdaXw5I'
' command is defined in the .mdp file
processing topology...
processing coordinates...
Program grompp, VERSION 3.3.1
Source code file: grompp.c, line: 443
Fatal error:
number of coordinates in coordinate file (fg.gro, 4057)
does not match topology (topol.top, 0)
will you please tell where i did wrong?
Thanks in advance
i finally used pdb2gmx -f filename.pdb -missing by default it will generate topol.top ,conf.gro and posre.itp files then i did
g_fg2cg -pfg topol.top -pcg system.top -c last_frame.pdb -n 0 -o fg.gro -wat 1
it will generate fg.gro file having coordinate. now i modify fg.mdp file and run further step i.e.,grompp -f fg.mdp -c fg.gro -p topol.top -o topol.tpr then it was showing error:
creating statusfile for 1 node...
' for variable integrator, using 'md'
Next time use one of: 'md' 'steep' 'cg' 'bd' 'sd' 'nm' 'l-bfgs' 'tpi'
' for variable ns-type, using 'Grid'
Next time use one of: 'Grid' 'Simple'
' for variable coulombtype, using 'Cut-off'ield
Next time use one of: 'Cut-off' 'Reaction-Field' 'Generalized-Reaction-Field' 'PM E' 'Ewald' 'PPPM' 'Poisson' 'Switch' 'Shift' 'User' 'Generalized-Born' 'Reaction- Field-nec' 'Encad-shift' 'PME-User'
' for variable tcoupl, using 'No'
Next time use one of: 'No' 'Berendsen' 'Nose-Hoover' 'yes' 'Andersen' 'Andersen-i nterval'
' for variable Pcoupl, using 'No'
Next time use one of: 'No' 'Berendsen' 'Parrinello-Rahman' 'Isotropic'
' for variable gen-vel, using 'no'
Next time use one of: 'no' 'yes'
' for variable constraints, using 'none'
Next time use one of: 'none' 'h-bonds' 'all-bonds' 'h-angles' 'all-angles'
' for variable rel_water, using 'no'
Next time use one of: 'no' 'yes'
checking input for internal consistency...
...ling /lib/cpp
: No such file or directory
cpp exit code: 32512
-I/usr/local/gromacs/share/gromacs/top topol.top > gromppdaXw5I'
' command is defined in the .mdp file
processing topology...
processing coordinates...
Program grompp, VERSION 3.3.1
Source code file: grompp.c, line: 443
Fatal error:
number of coordinates in coordinate file (fg.gro, 4057)
does not match topology (topol.top, 0)
will you please tell where i did wrong?
Thanks in advance
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
11 years 3 weeks ago #1528
by Clement
Replied by Clement on topic Reverse transformation
Could you copy/paste here the coordinates of the problematic glutamine? You said that you ran AA simulations of this protein with GROMACS 4.5.5 before... Which force field? Howmany residues does it have?
The topology you created using pdb2gmx is for the protein ONLY. In my previous post, I insisted on the difference between protein topology and system topology. You need to add the "[ system ]" and "[ molecules ]" sections to this topol.top and report there the number of water/ions/any constituent of your system.
Your .mdp file seems to be quite troublesome as well; a lot of parameters are just kept to default... You might want to check that.
The topology you created using pdb2gmx is for the protein ONLY. In my previous post, I insisted on the difference between protein topology and system topology. You need to add the "[ system ]" and "[ molecules ]" sections to this topol.top and report there the number of water/ions/any constituent of your system.
Your .mdp file seems to be quite troublesome as well; a lot of parameters are just kept to default... You might want to check that.
Please Log in or Create an account to join the conversation.
- nivedita
-
Topic Author
- Offline
- Junior Boarder
Less
More
- Posts: 36
11 years 2 weeks ago #1529
by nivedita
Replied by nivedita on topic Reverse transformation
hello clement,
my coordinate file for particular residue is
EXPDTA THEORETICAL MODEL, MODELLER 9.10 2012/06/21 17:51:18
REMARK 6 MODELLER OBJECTIVE FUNCTION: 3585.5024
REMARK 6 MODELLER BEST TEMPLATE % SEQ ID: 100.000
ATOM 1 N GLN 1 7.595 1.554 57.330 1.00 58.87 N
ATOM 2 CA GLN 1 6.376 1.170 56.584 1.00 58.87 C
ATOM 3 CB GLN 1 5.603 2.419 56.126 1.00 58.87 C
ATOM 4 CG GLN 1 4.895 3.150 57.268 1.00 58.87 C
ATOM 5 CD GLN 1 3.686 2.314 57.671 1.00 58.87 C
ATOM 6 OE1 GLN 1 2.542 2.728 57.493 1.00 58.87 O
ATOM 7 NE2 GLN 1 3.945 1.100 58.226 1.00 58.87 N
ATOM 8 C GLN 1 6.694 0.346 55.385 1.00 58.87 C
ATOM 9 O GLN 1 7.176 -0.780 55.502 1.00 58.87 O
and force field which i used for aa simulation is "gromos43a1.ff ".
my coordinate file for particular residue is
EXPDTA THEORETICAL MODEL, MODELLER 9.10 2012/06/21 17:51:18
REMARK 6 MODELLER OBJECTIVE FUNCTION: 3585.5024
REMARK 6 MODELLER BEST TEMPLATE % SEQ ID: 100.000
ATOM 1 N GLN 1 7.595 1.554 57.330 1.00 58.87 N
ATOM 2 CA GLN 1 6.376 1.170 56.584 1.00 58.87 C
ATOM 3 CB GLN 1 5.603 2.419 56.126 1.00 58.87 C
ATOM 4 CG GLN 1 4.895 3.150 57.268 1.00 58.87 C
ATOM 5 CD GLN 1 3.686 2.314 57.671 1.00 58.87 C
ATOM 6 OE1 GLN 1 2.542 2.728 57.493 1.00 58.87 O
ATOM 7 NE2 GLN 1 3.945 1.100 58.226 1.00 58.87 N
ATOM 8 C GLN 1 6.694 0.346 55.385 1.00 58.87 C
ATOM 9 O GLN 1 7.176 -0.780 55.502 1.00 58.87 O
and force field which i used for aa simulation is "gromos43a1.ff ".
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
11 years 2 weeks ago #1530
by Clement
Replied by Clement on topic Reverse transformation
The structure Modeller gave you doesn't seem to include hydrogens (even in united atom force fields the hydrogen from the peptidic backbone is explicitly modeled); but you simulated this protein with a GROMOS force field before: take the output of this simulation as your protein structure to create your topology. You won't have problems with hydrogens.
And you HAVE TO to use GROMOS 43a1 or 53a6 force fields when creating your protein topology with the tweaked version of GROMACS; the mapping and the extra dihedrals were added only to these two. It's explicitly written when you have to choose between the different force fields...
And you HAVE TO to use GROMOS 43a1 or 53a6 force fields when creating your protein topology with the tweaked version of GROMACS; the mapping and the extra dihedrals were added only to these two. It's explicitly written when you have to choose between the different force fields...
Please Log in or Create an account to join the conversation.
Time to create page: 0.125 seconds