Re: [gmx-users] g_rmsf fitting

[Tsjerk Wassenaar]

See http://wiki.gromacs.org/index.php/Periodic_Boundary_Conditions

On Jan 4, 2008 11:20 AM, tangxuan <[EMAIL PROTECTED]> wrote:

> Thanks for your detailed explanation. Sorry for my words and I wrote
> them in a hurry without careful check.  In fact, I can not see a whole
> protein in ref1.tpr and part of subunits are out of box.
> I do not know if this can be called jump.
>
> Tang
>
>
> Tsjerk Wassenaar wrote:
> > Hi Tang,
> >
> > The subunits have no contact each other obviously,
> > without jump in them and I can not see an intact protein.
> >
> >
> > Please be more clear and try to write full, correct sentences. I
> > suppose you mean that the subunits are separated at start, so there
> > has been a jump in the setup stage.
> >
> > However, the structure in the frames  shows that there is a whole
> > protein with jump.
> > Therefore, the structure in the tpr file is different from the
> > structure
> > in the first frame.
> >
> >
> > So (one of the) subunits occasionally jump(s) back and make the
> > molecule "whole".
> >
> > The simulation is 30ns long and I am trying to calculate the rmsf
> > between 20ns and 30ns. I removed the jumps in whole
> > simulation(trjconv
> > -f .xtc(.trr) -s ref1.tpr -pbc nojump) first and  got the nojump xtc
> > file between 20ns and 30ns.
> >
> >
> > Assuming that ref1.tpr corresponds to your starting structure, which
> > has the subunits separated, this means that you end up with a
> > trajectory which has the subunits separated consistently.
> >
> >
> > The main difference between is the tpr file
> > at 20ns(ref20.tpr) i used and two ways I have used to get it:
> > 1)  used ref1.tpr and original trr file to get a tpr file at 20ns
> > ref20.tpr by tpbconv.
> >
> >
> > This reference may be of the correct complex (check it).
> >
> > 2) used ref1.tpr and nojump trr file to get a tpr file at 20ns
> > ref20.tpr
> > by tpbconv.
> >
> >
> > And this will have the subunits separated.
> >
> >
> > then "g_rmsf -f .xtc (nojump) -s ref20.tpr   -res" is used to
> > calculate
> > the rmsf.
> >
> >
> > So the answer is in the structures (jump/nojump) and not in the way
> > g_rmsf performs the fit and calculates the rmsf. Try to make this line
> > of thinking your own and check the structures (references/selected
> > frames from trajectories) first whenever you encounter problems like
> > this. This issue has also been covered before, and I recall having
> > posted a message on this list regarding the correct use of references,
> > fitting and -pbc options using trjconv, which may be useful to you.
> >
> > Cheers,
> >
> > Tsjerk
> >
> >
> > --
> > Tsjerk A. Wassenaar, Ph.D.
> > Junior UD (post-doc)
> > Biomolecular NMR, Bijvoet Center
> > Utrecht University
> > Padualaan 8
> > 3584 CH Utrecht
> > The Netherlands
> > P: +31-30-2539931
> > F: +31-30-2537623
> > 
> >
> > ___
> > gmx-users mailing listgmx-users@gromacs.org
> > http://www.gromacs.org/mailman/listinfo/gmx-users
> > Please search the archive at http://www.gromacs.org/search before
> posting!
> > Please don't post (un)subscribe requests to the list. Use the
> > www interface or send it to [EMAIL PROTECTED]
> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>
> ___
> gmx-users mailing listgmx-users@gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to [EMAIL PROTECTED]
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>



-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623
___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [gmx-users] g_rmsf fitting

[tangxuan]

Thanks for your detailed explanation. Sorry for my words and I wrote 
them in a hurry without careful check.  In fact, I can not see a whole 
protein in ref1.tpr and part of subunits are out of box.

I do not know if this can be called jump.

Tang


Tsjerk Wassenaar wrote:

Hi Tang,

The subunits have no contact each other obviously,
without jump in them and I can not see an intact protein. 



Please be more clear and try to write full, correct sentences. I 
suppose you mean that the subunits are separated at start, so there 
has been a jump in the setup stage.


However, the structure in the frames  shows that there is a whole
protein with jump.
Therefore, the structure in the tpr file is different from the
structure
in the first frame.


So (one of the) subunits occasionally jump(s) back and make the 
molecule "whole".


The simulation is 30ns long and I am trying to calculate the rmsf
between 20ns and 30ns. I removed the jumps in whole
simulation(trjconv
-f .xtc(.trr) -s ref1.tpr -pbc nojump) first and  got the nojump xtc
file between 20ns and 30ns. 



Assuming that ref1.tpr corresponds to your starting structure, which 
has the subunits separated, this means that you end up with a 
trajectory which has the subunits separated consistently.
 


The main difference between is the tpr file
at 20ns(ref20.tpr) i used and two ways I have used to get it:
1)  used ref1.tpr and original trr file to get a tpr file at 20ns
ref20.tpr by tpbconv.


This reference may be of the correct complex (check it).

2) used ref1.tpr and nojump trr file to get a tpr file at 20ns
ref20.tpr
by tpbconv.


And this will have the subunits separated. 
 


then "g_rmsf -f .xtc (nojump) -s ref20.tpr   -res" is used to
calculate
the rmsf.

 
So the answer is in the structures (jump/nojump) and not in the way 
g_rmsf performs the fit and calculates the rmsf. Try to make this line 
of thinking your own and check the structures (references/selected 
frames from trajectories) first whenever you encounter problems like 
this. This issue has also been covered before, and I recall having 
posted a message on this list regarding the correct use of references, 
fitting and -pbc options using trjconv, which may be useful to you.


Cheers,

Tsjerk


--
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623



___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]

Can't post? Read http://www.gromacs.org/mailing_lists/users.php


___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]

Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [gmx-users] g_rmsf fitting

[Tsjerk Wassenaar]

Hi Tang,

The subunits have no contact each other obviously,
> without jump in them and I can not see an intact protein.


Please be more clear and try to write full, correct sentences. I suppose you
mean that the subunits are separated at start, so there has been a jump in
the setup stage.

However, the structure in the frames  shows that there is a whole protein
> with jump.
> Therefore, the structure in the tpr file is different from the structure
> in the first frame.
>

So (one of the) subunits occasionally jump(s) back and make the molecule
"whole".

The simulation is 30ns long and I am trying to calculate the rmsf
> between 20ns and 30ns. I removed the jumps in whole simulation(trjconv
> -f .xtc(.trr) -s ref1.tpr -pbc nojump) first and  got the nojump xtc
> file between 20ns and 30ns.


Assuming that ref1.tpr corresponds to your starting structure, which has the
subunits separated, this means that you end up with a trajectory which has
the subunits separated consistently.


> The main difference between is the tpr file
> at 20ns(ref20.tpr) i used and two ways I have used to get it:
> 1)  used ref1.tpr and original trr file to get a tpr file at 20ns
> ref20.tpr by tpbconv.


This reference may be of the correct complex (check it).

2) used ref1.tpr and nojump trr file to get a tpr file at 20ns ref20.tpr
> by tpbconv.


And this will have the subunits separated.


> then "g_rmsf -f .xtc (nojump) -s ref20.tpr   -res" is used to calculate
> the rmsf.
>

So the answer is in the structures (jump/nojump) and not in the way g_rmsf
performs the fit and calculates the rmsf. Try to make this line of thinking
your own and check the structures (references/selected frames from
trajectories) first whenever you encounter problems like this. This issue
has also been covered before, and I recall having posted a message on this
list regarding the correct use of references, fitting and -pbc options using
trjconv, which may be useful to you.

Cheers,

Tsjerk


-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623
___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [gmx-users] g_rmsf fitting

[David van der Spoel]

tangxuan wrote:

Hi Tsjerk,
Yes, you are right. My protein has multiple subunits.  I have a original 
tpr file ref1.tpr which is the input to start the simulation. I have 
checked the protein structure
in this tpr file. The subunits have no contact each other obviously, 
without jump in them and I can not see an intact protein. However, the 
structure in the frames  shows that there is a whole protein with jump.
Therefore, the structure in the tpr file is different from the structure 
in the first frame. The simulation is 30ns long and I am trying to 
calculate the rmsf between 20ns and 30ns. I removed the jumps in whole 
simulation(trjconv -f .xtc(.trr) -s ref1.tpr -pbc nojump) first and  got 
the nojump xtc file between 20ns and 30ns. The main difference between 
is the tpr file at 20ns(ref20.tpr) i used and two ways I have used to 
get it:
1)  used ref1.tpr and original trr file to get a tpr file at 20ns 
ref20.tpr by tpbconv.
2) used ref1.tpr and nojump trr file to get a tpr file at 20ns ref20.tpr 
by tpbconv.
then "g_rmsf -f .xtc (nojump) -s ref20.tpr   -res" is used to calculate 
the rmsf.
I take out part of the result below and the results are from two methods 
for same subunit.


What happens whan you overlay the two reference structures using 
g_confrms? Is it high RMSD?

Both reference structures and trajectories should be jump-free.
--
David.

David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  75124 Uppsala, Sweden
phone:  46 18 471 4205  fax: 46 18 511 755
[EMAIL PROTECTED]   [EMAIL PROTECTED]   http://folding.bmc.uu.se

___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]

Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [gmx-users] g_rmsf fitting

[tangxuan]

Hi Tsjerk,
Yes, you are right. My protein has multiple subunits.  I have a original 
tpr file ref1.tpr which is the input to start the simulation. I have 
checked the protein structure
in this tpr file. The subunits have no contact each other obviously, 
without jump in them and I can not see an intact protein. However, the 
structure in the frames  shows that there is a whole protein with jump.
Therefore, the structure in the tpr file is different from the structure 
in the first frame. 
The simulation is 30ns long and I am trying to calculate the rmsf 
between 20ns and 30ns. I removed the jumps in whole simulation(trjconv 
-f .xtc(.trr) -s ref1.tpr -pbc nojump) first and  got the nojump xtc 
file between 20ns and 30ns. The main difference between is the tpr file 
at 20ns(ref20.tpr) i used and two ways I have used to get it:
1)  used ref1.tpr and original trr file to get a tpr file at 20ns 
ref20.tpr by tpbconv.
2) used ref1.tpr and nojump trr file to get a tpr file at 20ns ref20.tpr 
by tpbconv.
then "g_rmsf -f .xtc (nojump) -s ref20.tpr   -res" is used to calculate 
the rmsf.
I take out part of the result below and the results are from two methods 
for same subunit.

1)
 62   0.1937
  63   0.2319
  64   0.2058
  65   0.2027
  66   0.1785
  67   0.1625
  68   0.1641
  69   0.1683
  70   0.1402
  71   0.1353
  72   0.1122
  73   0.0984
  74   0.0935
  75   0.1094
  76   0.0797
  77   0.1318
  78   0.0928
  79   0.1255
  80   0.1251
  81   0.1525
  82   0.1716
  83   0.1509
  84   0.1333
  85   0.0911
  86   0.0984
  87   0.0642
  88   0.0862
  89   0.0648
  90   0.0793
  91   0.0918
  92   0.1218
  93   0.1318
  94   0.1609
  95   0.1717
  96   0.1812
  97   0.1569
  98   0.1683
  99   0.1495 
 2)

 62   0.0843
  63   0.0946
  64   0.0801
  65   0.0652
  66   0.0801
  67   0.0605
  68   0.0728
  69   0.0733
  70   0.0561
  71   0.0559
  72   0.0628
  73   0.0552
  74   0.0613
  75   0.0758
  76   0.0639
  77   0.1172
  78   0.0677
  79   0.0877
  80   0.0981
  81   0.1313
  82   0.1490
  83   0.1460
  84   0.1133
  85   0.0947
  86   0.0761
  87   0.0602
  88   0.0677
  89   0.0587
  90   0.0530
  91   0.0545
  92   0.0647
  93   0.0451
  94   0.0560
  95   0.0541
  96   0.0572
  97   0.0486
  98   0.0535
  99   0.0459

Thanks

Tang


Tsjerk Wassenaar wrote:

Hi Tang,

It's just the general case of least-squares fitting:

1. Bring centres of geometry/mass to origin
2. Calculate (mass-weighted) rotation matrix
3. Rotate structure
(4. Calculate squared displacements)

But what do you mean with 'very different'. Can you provide an 
example? What command line did you use? Do you have multiple subunits 
and have jumps over PBC between the two reference structures?


Cheers,

Tsjerk

On Jan 3, 2008 3:37 PM, tangxuan <[EMAIL PROTECTED] 
> wrote:


Hi,
Does anyone know how  fitting step works in detail when running
g_rmsf?
When I used two different tpr files with same protein structure but
different atoms position, the result is very different.
Can you explain this ?
many thanks!

Tang
___
gmx-users mailing listgmx-users@gromacs.org

http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before
posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to [EMAIL PROTECTED]
.
Can't post? Read http://www.gromacs.org/mailing_lists/users.php





--
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623



___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]

Can't post? Read http://www.gromacs.org/mailing_lists/users.php


___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]

Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [gmx-users] g_rmsf fitting

[Tsjerk Wassenaar]

Hi Tang,

It's just the general case of least-squares fitting:

1. Bring centres of geometry/mass to origin
2. Calculate (mass-weighted) rotation matrix
3. Rotate structure
(4. Calculate squared displacements)

But what do you mean with 'very different'. Can you provide an example? What
command line did you use? Do you have multiple subunits and have jumps over
PBC between the two reference structures?

Cheers,

Tsjerk

On Jan 3, 2008 3:37 PM, tangxuan <[EMAIL PROTECTED]> wrote:

> Hi,
> Does anyone know how  fitting step works in detail when running g_rmsf?
> When I used two different tpr files with same protein structure but
> different atoms position, the result is very different.
> Can you explain this ?
> many thanks!
>
> Tang
> ___
> gmx-users mailing listgmx-users@gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to [EMAIL PROTECTED]
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>



-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623
___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

[gmx-users] g_rmsf fitting

[tangxuan]

Hi,
Does anyone know how  fitting step works in detail when running g_rmsf? 
When I used two different tpr files with same protein structure but 
different atoms position, the result is very different.
Can you explain this ? 
many thanks!


Tang
___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]

Can't post? Read http://www.gromacs.org/mailing_lists/users.php