Re: [gmx-users] shift function for vdwtype and density

2009-05-19 Thread Mark Abraham 

Yanmei Song wrote:

Dear all:

This is the mdp I just used for the system. The funny thing is after a 
8ns simulation I got a very high pressure and temperature:


 Temperature   343.437   
Pressure (bar)   45.782 

I also did another run with removing the vdw_switch in mdp and keep all 
other condition unchanged.  I got good result.


The following is my mdp:

cpp =  /lib/cpp
constraints =  all_bonds
integrator  =  md
dt  =  0.004   ; ps !


I'd suspect that this value for dt is also not likely to have been used 
during parameterization.


Mark
___
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 gmx-users-requ...@gromacs.org.

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

Re: [gmx-users] shift function for vdwtype and density

2009-05-19 Thread Yanmei Song 
Dear all:

This is the mdp I just used for the system. The funny thing is after a 8ns
simulation I got a very high pressure and temperature:

 Temperature   343.437
Pressure (bar)   45.782

I also did another run with removing the vdw_switch in mdp and keep all
other condition unchanged.  I got good result.

The following is my mdp:

cpp =  /lib/cpp
constraints =  all_bonds
integrator  =  md
dt  =  0.004   ; ps !
nsteps  =  2000   ; total 8ns.
nstcomm =  1
nstxout =  5
nstvout =  5
nstfout =  0
nstlog  =  5000
nstenergy   =  5000
nstxtcout   =  25000
nstlist =  10
ns_type =  grid
pbc =  xyz
coulombtype =  PME
vdwtype =  Shift
rvdw_switch =  0.8
rlist   =  1.2
rcoulomb=  1.2
rvdw=  1.0
fourierspacing  =  0.12
pme_order   =  4
ewald_rtol  =  1e-5
; Berendsen temperature coupling is on in two groups
Tcoupl  =  berendsen
tc_grps =  PDM
tau_t   =  0.1
ref_t   =  300
; Energy monitoring
energygrps  =  PDM
; Isotropic pressure coupling is now on
Pcoupl  =  berendsen
pcoupltype  =  isotropic
;pc-grps=  PDM
tau_p   =  1.0


On Tue, May 19, 2009 at 11:12 AM, Yanmei Song  wrote:

> Dear David:
>
> Thank you so much for the valuable information.
>
> "Our conclusion in the above paper is to use PME for Coulomb and not shift
> Van der Waals"
>
> I have read the paper you mentioned and did not find how you treated the
> VdW. Have you done some work on the cut off method for vdw? Since I used PME
> for Coulomb in my system. I am still confusing why not shift for vdw.
>
> " I would also recommend to use TIP4P rather than TIP5P, as Jorgensen uses
> TIP4P for all his parameterizations."
>
> I have read Jorgensen's paper for the OPLS and realized that the TIP4P
> would be the first choice for combining with OPLS. However my concern is
> that TIP4P model is not as good as TIP5P in terms of diffusion constant and
> viscosity which are extremely important for my purpose.
>
> I also have compared SPC and TIP4P and turn out that SPC can give a better
> result for my system. Unfornately SPC has a very bad viscosity as TIP4P. So
> I am trying TIP5P.
>
> Do you think OPLS+ TIP5P can cause serious problem or horrible artifacts.
> If you can suggest some literatures on this that would be great.
>
> I really appericate all your help!
>
> On Mon, May 18, 2009 at 10:37 PM, David van der Spoel <
> sp...@xray.bmc.uu.se> wrote:
>
>> Justin A. Lemkul wrote:
>>
>>>
>>>
>>> Yanmei Song wrote:
>>>
 Dear Justin:

 Yes and I tried dispersion correction and it gives a larger density than
 I expected.  The density of the polymer by vdwtype=shift is more close to
 the experimental value. Then your suggestion would be to choose the
 vdwtype=cut-off under this condition even if it result in larger density?
 Thank you so much!


>>> No.  My suggestion is to do your homework, as Mark originally said, and
>>> as I echoed in my last mail.  Read about the water model, its expected
>>> performance under different simulation parameters, and evaluate what you
>>> want to do once you have done this background work.  If you have done a
>>> thorough literature review, there is no one else that can make a choice for
>>> you; decide how to handle your system based on what you read.
>>>
>>>  Actually this is not so easy. Jorgensen has never published the way he
>> treats cut-offs, but he was friendly enough to let me know, so we have
>> actually done it:
>> David van der Spoel and Paul J. van Maaren: The origin of layer structure
>> artifacts in simulation of liquid water J. Chem. Theor. Comp. 2 pp. 1-11
>> (2006)
>>
>> If you look in the appendix you can find the "one and only"
>
>
>
>> OPLS. However This is not to be recommended since the effect is very
>> different for MD and MC (what Jorgensen uses). In MD the official method
>> will give horrible artifacts. In other words, you have to use something
>> better. Our conclusion in the above paper is to use PME for Coulomb and not
>> shift Van der Waals. I would also recommend to use TIP4P rather than TIP5P,
>> as Jorgensen uses TIP4P for all his parameterizations.
>>
>>
>>
>>
>>  -Justin
>>>
>>>  On Mon, May 18, 2009 at 4:29 PM, Justin A. Lemkul >>> jalem...@vt.edu>> wrote:



Yanmei Song wrote:

Dear Justin:

Yes. Before I do the water+polymer, I did both the bulk systems
separately.  The density of TIP5P I mentioned earlier is the
results for pure water bulk system. My concern is that if it is
reasonable that the property of the system changes with
different mdp conditions? Since whe

Re: [gmx-users] shift function for vdwtype and density

2009-05-19 Thread Yanmei Song 
Dear David:

Thank you so much for the valuable information.

"Our conclusion in the above paper is to use PME for Coulomb and not shift
Van der Waals"

I have read the paper you mentioned and did not find how you treated the
VdW. Have you done some work on the cut off method for vdw? Since I used PME
for Coulomb in my system. I am still confusing why not shift for vdw.

" I would also recommend to use TIP4P rather than TIP5P, as Jorgensen uses
TIP4P for all his parameterizations."

I have read Jorgensen's paper for the OPLS and realized that the TIP4P would
be the first choice for combining with OPLS. However my concern is that
TIP4P model is not as good as TIP5P in terms of diffusion constant and
viscosity which are extremely important for my purpose.

I also have compared SPC and TIP4P and turn out that SPC can give a better
result for my system. Unfornately SPC has a very bad viscosity as TIP4P. So
I am trying TIP5P.

Do you think OPLS+ TIP5P can cause serious problem or horrible artifacts. If
you can suggest some literatures on this that would be great.

I really appericate all your help!

On Mon, May 18, 2009 at 10:37 PM, David van der Spoel
wrote:

> Justin A. Lemkul wrote:
>
>>
>>
>> Yanmei Song wrote:
>>
>>> Dear Justin:
>>>
>>> Yes and I tried dispersion correction and it gives a larger density than
>>> I expected.  The density of the polymer by vdwtype=shift is more close to
>>> the experimental value. Then your suggestion would be to choose the
>>> vdwtype=cut-off under this condition even if it result in larger density?
>>> Thank you so much!
>>>
>>>
>> No.  My suggestion is to do your homework, as Mark originally said, and as
>> I echoed in my last mail.  Read about the water model, its expected
>> performance under different simulation parameters, and evaluate what you
>> want to do once you have done this background work.  If you have done a
>> thorough literature review, there is no one else that can make a choice for
>> you; decide how to handle your system based on what you read.
>>
>>  Actually this is not so easy. Jorgensen has never published the way he
> treats cut-offs, but he was friendly enough to let me know, so we have
> actually done it:
> David van der Spoel and Paul J. van Maaren: The origin of layer structure
> artifacts in simulation of liquid water J. Chem. Theor. Comp. 2 pp. 1-11
> (2006)
>
> If you look in the appendix you can find the "one and only"



> OPLS. However This is not to be recommended since the effect is very
> different for MD and MC (what Jorgensen uses). In MD the official method
> will give horrible artifacts. In other words, you have to use something
> better. Our conclusion in the above paper is to use PME for Coulomb and not
> shift Van der Waals. I would also recommend to use TIP4P rather than TIP5P,
> as Jorgensen uses TIP4P for all his parameterizations.
>
>
>
>
>  -Justin
>>
>>  On Mon, May 18, 2009 at 4:29 PM, Justin A. Lemkul >> jalem...@vt.edu>> wrote:
>>>
>>>
>>>
>>>Yanmei Song wrote:
>>>
>>>Dear Justin:
>>>
>>>Yes. Before I do the water+polymer, I did both the bulk systems
>>>separately.  The density of TIP5P I mentioned earlier is the
>>>results for pure water bulk system. My concern is that if it is
>>>reasonable that the property of the system changes with
>>>different mdp conditions? Since when I did the bulk water and
>>>bulk polymer system, I found the densities change with different
>>>cutoff and vdwtype conditions. Is this a artifact? Or the larger
>>>cutoff, the more accurate result we can get?
>>>
>>>
>>>Every parameter you specify will affect the results you see.  I
>>>agree with Mark - look up the original reference for TIP5P and see
>>>what parameters are recommended.  Also read the subsequent
>>>literature that has used TIP5P to see what modifications
>>>(electrostatic treatment, vdW treatment, etc) are appropriate or
>>>even more accurate.
>>>
>>>Also realize that using vdwtype = shift may not be the best
>>>solution.  I only suggested as a guess for something to try back
>>>when you were getting bizarre results with your polymer.  Have you
>>>ever tried dispersion correction when using vdwtype = cutoff?  I
>>>routinely use such a setup and have never seen any really bizarre
>>>behavior.
>>>
>>>-Justin
>>>
>>>On Mon, May 18, 2009 at 4:10 PM, Justin A. Lemkul
>>>mailto:jalem...@vt.edu>
>>>>> wrote:
>>>
>>>
>>>
>>>   Yanmei Song wrote:
>>>
>>>   Dear Mark:
>>>
>>>   Thanks for your answer. I have to use the same mdp file
>>>for both
>>>   water and another polymer molecules in my system, right?
>>> Then
>>>   if the shift function works well for the polymer, I can
>>>not get
>>>   the best result for TIP5P?  The density of TIP5P is
>>> di

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread David van der Spoel 

Justin A. Lemkul wrote:



Yanmei Song wrote:

Dear Justin:

Yes and I tried dispersion correction and it gives a larger density 
than I expected.  The density of the polymer by vdwtype=shift is more 
close to the experimental value. Then your suggestion would be to 
choose the vdwtype=cut-off under this condition even if it result in 
larger density? Thank you so much!




No.  My suggestion is to do your homework, as Mark originally said, and 
as I echoed in my last mail.  Read about the water model, its expected 
performance under different simulation parameters, and evaluate what you 
want to do once you have done this background work.  If you have done a 
thorough literature review, there is no one else that can make a choice 
for you; decide how to handle your system based on what you read.


Actually this is not so easy. Jorgensen has never published the way he 
treats cut-offs, but he was friendly enough to let me know, so we have 
actually done it:
David van der Spoel and Paul J. van Maaren: The origin of layer 
structure artifacts in simulation of liquid water J. Chem. Theor. Comp. 
2 pp. 1-11 (2006)


If you look in the appendix you can find the "one and only" shifting 
method for OPLS. However This is not to be recommended since the 
effect is very different for MD and MC (what Jorgensen uses). In MD the 
official method will give horrible artifacts. In other words, you have 
to use something better. Our conclusion in the above paper is to use PME 
for Coulomb and not shift Van der Waals. I would also recommend to use 
TIP4P rather than TIP5P, as Jorgensen uses TIP4P for all his 
parameterizations.





-Justin

On Mon, May 18, 2009 at 4:29 PM, Justin A. Lemkul > wrote:




Yanmei Song wrote:

Dear Justin:

Yes. Before I do the water+polymer, I did both the bulk systems
separately.  The density of TIP5P I mentioned earlier is the
results for pure water bulk system. My concern is that if it is
reasonable that the property of the system changes with
different mdp conditions? Since when I did the bulk water and
bulk polymer system, I found the densities change with different
cutoff and vdwtype conditions. Is this a artifact? Or the larger
cutoff, the more accurate result we can get?


Every parameter you specify will affect the results you see.  I
agree with Mark - look up the original reference for TIP5P and see
what parameters are recommended.  Also read the subsequent
literature that has used TIP5P to see what modifications
(electrostatic treatment, vdW treatment, etc) are appropriate or
even more accurate.

Also realize that using vdwtype = shift may not be the best
solution.  I only suggested as a guess for something to try back
when you were getting bizarre results with your polymer.  Have you
ever tried dispersion correction when using vdwtype = cutoff?  I
routinely use such a setup and have never seen any really bizarre
behavior.

-Justin

On Mon, May 18, 2009 at 4:10 PM, Justin A. Lemkul
mailto:jalem...@vt.edu>
>> wrote:



   Yanmei Song wrote:

   Dear Mark:

   Thanks for your answer. I have to use the same mdp file
for both
   water and another polymer molecules in my system, right?
 Then
   if the shift function works well for the polymer, I can
not get
   the best result for TIP5P?  The density of TIP5P is 
different

   from the literature value of 982. Thanks!


   So your system is water + polymer?  In that case, you have an
added
   variable - the interaction between the water and your
polymer.  What
   you should probably be more concerned with the what the bulk
density
   of the whole system is.  I have found that analyzing 
densities of

   portions of mixed systems does not give an accurate result.

   -Justin


   On Mon, May 18, 2009 at 3:33 PM, Mark Abraham
   mailto:mark.abra...@anu.edu.au>
>

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Yanmei Song 
Dear Justin:

Thanks so much for your effort and information. Perhaps I did not set the
proper value for vdw_switch. I will try some value. I really appreciate your
nice help!

On Mon, May 18, 2009 at 5:24 PM, Justin A. Lemkul  wrote:

>
> Perhaps I can contribute something else that might be a bit more useful.
>  It just occurred to me that if you are only changing the vdwtype parameter,
> then you are not correctly implementing the shift function.  See here for
> proper use of a shift potential:
>
> http://www.gromacs.org/documentation/reference_4.0/online/mdp_opt.html#vdw
>
> Take note of the necessary changes to vdw_shift, and potentially rlist.  A
> few quick tests of the shift function indicate the following:
>
> Case 1. Plain cutoff's at 1.2 nm
> Density: 988 kg/m^3
>
> Case 2. Shift potential (no other changes, i.e. rvdw_switch = 0, rvdw =
> 1.2)
> Density: 962 kg/m^3
>
> Case 3. More appropriate shift potential (rvdw_switch = 1.0, rvdw = 1.2)
> Density: 983 kg/m^3
>
> -Justin
>
>
> Justin A. Lemkul wrote:
>
>>
>>
>> Yanmei Song wrote:
>>
>>> Dear Justin:
>>>
>>> Yes and I tried dispersion correction and it gives a larger density than
>>> I expected.  The density of the polymer by vdwtype=shift is more close to
>>> the experimental value. Then your suggestion would be to choose the
>>> vdwtype=cut-off under this condition even if it result in larger density?
>>> Thank you so much!
>>>
>>>
>> No.  My suggestion is to do your homework, as Mark originally said, and as
>> I echoed in my last mail.  Read about the water model, its expected
>> performance under different simulation parameters, and evaluate what you
>> want to do once you have done this background work.  If you have done a
>> thorough literature review, there is no one else that can make a choice for
>> you; decide how to handle your system based on what you read.
>>
>> -Justin
>>
>>  On Mon, May 18, 2009 at 4:29 PM, Justin A. Lemkul >> jalem...@vt.edu>> wrote:
>>>
>>>
>>>
>>>Yanmei Song wrote:
>>>
>>>Dear Justin:
>>>
>>>Yes. Before I do the water+polymer, I did both the bulk systems
>>>separately.  The density of TIP5P I mentioned earlier is the
>>>results for pure water bulk system. My concern is that if it is
>>>reasonable that the property of the system changes with
>>>different mdp conditions? Since when I did the bulk water and
>>>bulk polymer system, I found the densities change with different
>>>cutoff and vdwtype conditions. Is this a artifact? Or the larger
>>>cutoff, the more accurate result we can get?
>>>
>>>
>>>Every parameter you specify will affect the results you see.  I
>>>agree with Mark - look up the original reference for TIP5P and see
>>>what parameters are recommended.  Also read the subsequent
>>>literature that has used TIP5P to see what modifications
>>>(electrostatic treatment, vdW treatment, etc) are appropriate or
>>>even more accurate.
>>>
>>>Also realize that using vdwtype = shift may not be the best
>>>solution.  I only suggested as a guess for something to try back
>>>when you were getting bizarre results with your polymer.  Have you
>>>ever tried dispersion correction when using vdwtype = cutoff?  I
>>>routinely use such a setup and have never seen any really bizarre
>>>behavior.
>>>
>>>-Justin
>>>
>>>On Mon, May 18, 2009 at 4:10 PM, Justin A. Lemkul
>>>mailto:jalem...@vt.edu>
>>>>> wrote:
>>>
>>>
>>>
>>>   Yanmei Song wrote:
>>>
>>>   Dear Mark:
>>>
>>>   Thanks for your answer. I have to use the same mdp file
>>>for both
>>>   water and another polymer molecules in my system, right?
>>> Then
>>>   if the shift function works well for the polymer, I can
>>>not get
>>>   the best result for TIP5P?  The density of TIP5P is
>>> different
>>>   from the literature value of 982. Thanks!
>>>
>>>
>>>   So your system is water + polymer?  In that case, you have an
>>>added
>>>   variable - the interaction between the water and your
>>>polymer.  What
>>>   you should probably be more concerned with the what the bulk
>>>density
>>>   of the whole system is.  I have found that analyzing densities
>>> of
>>>   portions of mixed systems does not give an accurate result.
>>>
>>>   -Justin
>>>
>>>
>>>   On Mon, May 18, 2009 at 3:33 PM, Mark Abraham
>>>   mailto:mark.abra...@anu.edu.au>
>>>>
>>>   >>
>>>   >>>>
>>>  Yanmei Song wrote:
>>>
>>>

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Justin A. Lemkul 


Perhaps I can contribute something else that might be a bit more useful.  It 
just occurred to me that if you are only changing the vdwtype parameter, then 
you are not correctly implementing the shift function.  See here for proper use 
of a shift potential:


http://www.gromacs.org/documentation/reference_4.0/online/mdp_opt.html#vdw

Take note of the necessary changes to vdw_shift, and potentially rlist.  A few 
quick tests of the shift function indicate the following:


Case 1. Plain cutoff's at 1.2 nm
Density: 988 kg/m^3

Case 2. Shift potential (no other changes, i.e. rvdw_switch = 0, rvdw = 1.2)
Density: 962 kg/m^3

Case 3. More appropriate shift potential (rvdw_switch = 1.0, rvdw = 1.2)
Density: 983 kg/m^3

-Justin

Justin A. Lemkul wrote:



Yanmei Song wrote:

Dear Justin:

Yes and I tried dispersion correction and it gives a larger density 
than I expected.  The density of the polymer by vdwtype=shift is more 
close to the experimental value. Then your suggestion would be to 
choose the vdwtype=cut-off under this condition even if it result in 
larger density? Thank you so much!




No.  My suggestion is to do your homework, as Mark originally said, and 
as I echoed in my last mail.  Read about the water model, its expected 
performance under different simulation parameters, and evaluate what you 
want to do once you have done this background work.  If you have done a 
thorough literature review, there is no one else that can make a choice 
for you; decide how to handle your system based on what you read.


-Justin

On Mon, May 18, 2009 at 4:29 PM, Justin A. Lemkul > wrote:




Yanmei Song wrote:

Dear Justin:

Yes. Before I do the water+polymer, I did both the bulk systems
separately.  The density of TIP5P I mentioned earlier is the
results for pure water bulk system. My concern is that if it is
reasonable that the property of the system changes with
different mdp conditions? Since when I did the bulk water and
bulk polymer system, I found the densities change with different
cutoff and vdwtype conditions. Is this a artifact? Or the larger
cutoff, the more accurate result we can get?


Every parameter you specify will affect the results you see.  I
agree with Mark - look up the original reference for TIP5P and see
what parameters are recommended.  Also read the subsequent
literature that has used TIP5P to see what modifications
(electrostatic treatment, vdW treatment, etc) are appropriate or
even more accurate.

Also realize that using vdwtype = shift may not be the best
solution.  I only suggested as a guess for something to try back
when you were getting bizarre results with your polymer.  Have you
ever tried dispersion correction when using vdwtype = cutoff?  I
routinely use such a setup and have never seen any really bizarre
behavior.

-Justin

On Mon, May 18, 2009 at 4:10 PM, Justin A. Lemkul
mailto:jalem...@vt.edu>
>> wrote:



   Yanmei Song wrote:

   Dear Mark:

   Thanks for your answer. I have to use the same mdp file
for both
   water and another polymer molecules in my system, right?
 Then
   if the shift function works well for the polymer, I can
not get
   the best result for TIP5P?  The density of TIP5P is 
different

   from the literature value of 982. Thanks!


   So your system is water + polymer?  In that case, you have an
added
   variable - the interaction between the water and your
polymer.  What
   you should probably be more concerned with the what the bulk
density
   of the whole system is.  I have found that analyzing 
densities of

   portions of mixed systems does not give an accurate result.

   -Justin


   On Mon, May 18, 2009 at 3:33 PM, Mark Abraham
   mailto:mark.abra...@anu.edu.au>
>

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Justin A. Lemkul 



Yanmei Song wrote:

Dear Justin:

Yes and I tried dispersion correction and it gives a larger density than 
I expected.  The density of the polymer by vdwtype=shift is more close 
to the experimental value. Then your suggestion would be to choose the 
vdwtype=cut-off under this condition even if it result in larger 
density? Thank you so much!




No.  My suggestion is to do your homework, as Mark originally said, and as I 
echoed in my last mail.  Read about the water model, its expected performance 
under different simulation parameters, and evaluate what you want to do once you 
have done this background work.  If you have done a thorough literature review, 
there is no one else that can make a choice for you; decide how to handle your 
system based on what you read.


-Justin

On Mon, May 18, 2009 at 4:29 PM, Justin A. Lemkul > wrote:




Yanmei Song wrote:

Dear Justin:

Yes. Before I do the water+polymer, I did both the bulk systems
separately.  The density of TIP5P I mentioned earlier is the
results for pure water bulk system. My concern is that if it is
reasonable that the property of the system changes with
different mdp conditions? Since when I did the bulk water and
bulk polymer system, I found the densities change with different
cutoff and vdwtype conditions. Is this a artifact? Or the larger
cutoff, the more accurate result we can get?


Every parameter you specify will affect the results you see.  I
agree with Mark - look up the original reference for TIP5P and see
what parameters are recommended.  Also read the subsequent
literature that has used TIP5P to see what modifications
(electrostatic treatment, vdW treatment, etc) are appropriate or
even more accurate.

Also realize that using vdwtype = shift may not be the best
solution.  I only suggested as a guess for something to try back
when you were getting bizarre results with your polymer.  Have you
ever tried dispersion correction when using vdwtype = cutoff?  I
routinely use such a setup and have never seen any really bizarre
behavior.

-Justin

On Mon, May 18, 2009 at 4:10 PM, Justin A. Lemkul
mailto:jalem...@vt.edu>
>> wrote:



   Yanmei Song wrote:

   Dear Mark:

   Thanks for your answer. I have to use the same mdp file
for both
   water and another polymer molecules in my system, right?
 Then
   if the shift function works well for the polymer, I can
not get
   the best result for TIP5P?  The density of TIP5P is different
   from the literature value of 982. Thanks!


   So your system is water + polymer?  In that case, you have an
added
   variable - the interaction between the water and your
polymer.  What
   you should probably be more concerned with the what the bulk
density
   of the whole system is.  I have found that analyzing densities of
   portions of mixed systems does not give an accurate result.

   -Justin


   On Mon, May 18, 2009 at 3:33 PM, Mark Abraham
   mailto:mark.abra...@anu.edu.au>
>
   
   
   >
   >>


  http://www.gromacs.org/mailman/listinfo/gmx-users
  Please search the a

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Yanmei Song 
Dear Justin:

Yes and I tried dispersion correction and it gives a larger density than I
expected.  The density of the polymer by vdwtype=shift is more close to the
experimental value. Then your suggestion would be to choose the
vdwtype=cut-off under this condition even if it result in larger density?
Thank you so much!

On Mon, May 18, 2009 at 4:29 PM, Justin A. Lemkul  wrote:

>
>
> Yanmei Song wrote:
>
>> Dear Justin:
>>
>> Yes. Before I do the water+polymer, I did both the bulk systems
>> separately.  The density of TIP5P I mentioned earlier is the results for
>> pure water bulk system. My concern is that if it is reasonable that the
>> property of the system changes with different mdp conditions? Since when I
>> did the bulk water and bulk polymer system, I found the densities change
>> with different cutoff and vdwtype conditions. Is this a artifact? Or the
>> larger cutoff, the more accurate result we can get?
>>
>>
> Every parameter you specify will affect the results you see.  I agree with
> Mark - look up the original reference for TIP5P and see what parameters are
> recommended.  Also read the subsequent literature that has used TIP5P to see
> what modifications (electrostatic treatment, vdW treatment, etc) are
> appropriate or even more accurate.
>
> Also realize that using vdwtype = shift may not be the best solution.  I
> only suggested as a guess for something to try back when you were getting
> bizarre results with your polymer.  Have you ever tried dispersion
> correction when using vdwtype = cutoff?  I routinely use such a setup and
> have never seen any really bizarre behavior.
>
> -Justin
>
>  On Mon, May 18, 2009 at 4:10 PM, Justin A. Lemkul > jalem...@vt.edu>> wrote:
>>
>>
>>
>>Yanmei Song wrote:
>>
>>Dear Mark:
>>
>>Thanks for your answer. I have to use the same mdp file for both
>>water and another polymer molecules in my system, right?  Then
>>if the shift function works well for the polymer, I can not get
>>the best result for TIP5P?  The density of TIP5P is different
>>from the literature value of 982. Thanks!
>>
>>
>>So your system is water + polymer?  In that case, you have an added
>>variable - the interaction between the water and your polymer.  What
>>you should probably be more concerned with the what the bulk density
>>of the whole system is.  I have found that analyzing densities of
>>portions of mixed systems does not give an accurate result.
>>
>>-Justin
>>
>>
>>On Mon, May 18, 2009 at 3:33 PM, Mark Abraham
>>mailto:mark.abra...@anu.edu.au>
>>>>> wrote:
>>
>>   Yanmei Song wrote:
>>
>>   Dear All:
>>
>>   I found when I changed the vdwtype from cut-off to shift,
>> the
>>   density of the system decrease significantly.All the other
>>   setting in the mdp file remain unchanged. For example the
>>   density of TIP5P water drops to 962. Does it mean I can
>>not use
>>   the shift for vdwtype?
>>
>>
>>   Maybe not. Check out the paper in which TIP5P was
>>parameterized. The
>>   greater the extent to which your .mdp conditions differ from
>>those
>>   conditions, the greater the likelihood that the
>>parameterization is
>>   no longer valid.
>>
>>   Mark
>>   ___
>>   gmx-users mailing listgmx-users@gromacs.org
>>
>>   >
>>
>>   http://www.gromacs.org/mailman/listinfo/gmx-users
>>   Please search the archive at http://www.gromacs.org/searchbefore
>>   posting!
>>   Please don't post (un)subscribe requests to the list. Use the
>> www
>>   interface or send it to gmx-users-requ...@gromacs.org
>>
>>   >>.
>>
>>   Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>
>>
>>
>>
>>--Yanmei Song
>>Ph.D. Candidate
>>Department of Chemical Engineering
>>Arizona State University
>>
>>
>>
>>  
>>
>>
>>___
>>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 gmx-users-requ...@gromacs.org
>>.
>

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Justin A. Lemkul 



Yanmei Song wrote:

Dear Justin:

Yes. Before I do the water+polymer, I did both the bulk systems 
separately.  The density of TIP5P I mentioned earlier is the results for 
pure water bulk system. My concern is that if it is reasonable that the 
property of the system changes with different mdp conditions? Since when 
I did the bulk water and bulk polymer system, I found the densities 
change with different cutoff and vdwtype conditions. Is this a artifact? 
Or the larger cutoff, the more accurate result we can get?




Every parameter you specify will affect the results you see.  I agree with Mark 
- look up the original reference for TIP5P and see what parameters are 
recommended.  Also read the subsequent literature that has used TIP5P to see 
what modifications (electrostatic treatment, vdW treatment, etc) are appropriate 
or even more accurate.


Also realize that using vdwtype = shift may not be the best solution.  I only 
suggested as a guess for something to try back when you were getting bizarre 
results with your polymer.  Have you ever tried dispersion correction when using 
vdwtype = cutoff?  I routinely use such a setup and have never seen any really 
bizarre behavior.


-Justin

On Mon, May 18, 2009 at 4:10 PM, Justin A. Lemkul > wrote:




Yanmei Song wrote:

Dear Mark:

Thanks for your answer. I have to use the same mdp file for both
water and another polymer molecules in my system, right?  Then
if the shift function works well for the polymer, I can not get
the best result for TIP5P?  The density of TIP5P is different
from the literature value of 982. Thanks!


So your system is water + polymer?  In that case, you have an added
variable - the interaction between the water and your polymer.  What
you should probably be more concerned with the what the bulk density
of the whole system is.  I have found that analyzing densities of
portions of mixed systems does not give an accurate result.

-Justin


On Mon, May 18, 2009 at 3:33 PM, Mark Abraham
mailto:mark.abra...@anu.edu.au>
>> wrote:

   Yanmei Song wrote:

   Dear All:

   I found when I changed the vdwtype from cut-off to shift, the
   density of the system decrease significantly.All the other
   setting in the mdp file remain unchanged. For example the
   density of TIP5P water drops to 962. Does it mean I can
not use
   the shift for vdwtype?


   Maybe not. Check out the paper in which TIP5P was
parameterized. The
   greater the extent to which your .mdp conditions differ from
those
   conditions, the greater the likelihood that the
parameterization is
   no longer valid.

   Mark
   ___
   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 gmx-users-requ...@gromacs.org

   >.

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




-- 
Yanmei Song

Ph.D. Candidate
Department of Chemical Engineering
Arizona State University





___
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 gmx-users-requ...@gromacs.org
.
Can't post? Read http://www.gromacs.org/mailing_lists/users.php


-- 



Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu  | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin



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

http://www.gromacs.

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Yanmei Song 
Dear Justin:

Yes. Before I do the water+polymer, I did both the bulk systems separately.
The density of TIP5P I mentioned earlier is the results for pure water bulk
system. My concern is that if it is reasonable that the property of the
system changes with different mdp conditions? Since when I did the bulk
water and bulk polymer system, I found the densities change with different
cutoff and vdwtype conditions. Is this a artifact? Or the larger cutoff, the
more accurate result we can get?

On Mon, May 18, 2009 at 4:10 PM, Justin A. Lemkul  wrote:

>
>
> Yanmei Song wrote:
>
>> Dear Mark:
>>
>> Thanks for your answer. I have to use the same mdp file for both water and
>> another polymer molecules in my system, right?  Then if the shift function
>> works well for the polymer, I can not get the best result for TIP5P?  The
>> density of TIP5P is different from the literature value of 982. Thanks!
>>
>
> So your system is water + polymer?  In that case, you have an added
> variable - the interaction between the water and your polymer.  What you
> should probably be more concerned with the what the bulk density of the
> whole system is.  I have found that analyzing densities of portions of mixed
> systems does not give an accurate result.
>
> -Justin
>
>
>> On Mon, May 18, 2009 at 3:33 PM, Mark Abraham 
>> > mark.abra...@anu.edu.au>> wrote:
>>
>>Yanmei Song wrote:
>>
>>Dear All:
>>
>>I found when I changed the vdwtype from cut-off to shift, the
>>density of the system decrease significantly.All the other
>>setting in the mdp file remain unchanged. For example the
>>density of TIP5P water drops to 962. Does it mean I can not use
>>the shift for vdwtype?
>>
>>
>>Maybe not. Check out the paper in which TIP5P was parameterized. The
>>greater the extent to which your .mdp conditions differ from those
>>conditions, the greater the likelihood that the parameterization is
>>no longer valid.
>>
>>Mark
>>___
>>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 gmx-users-requ...@gromacs.org
>>.
>>Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>
>>
>>
>>
>> --
>> Yanmei Song
>> Ph.D. Candidate
>> Department of Chemical Engineering
>> Arizona State University
>>
>>
>> 
>>
>> ___
>> 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 gmx-users-requ...@gromacs.org.
>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>
>
> --
> 
>
> Justin A. Lemkul
> Ph.D. Candidate
> ICTAS Doctoral Scholar
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>
> 
>
> ___
> 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 gmx-users-requ...@gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>



-- 
Yanmei Song
Ph.D. Candidate
Department of Chemical Engineering
Arizona State University
___
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 gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Justin A. Lemkul 



Yanmei Song wrote:

Dear Mark:

Thanks for your answer. I have to use the same mdp file for both water 
and another polymer molecules in my system, right?  Then if the shift 
function works well for the polymer, I can not get the best result for 
TIP5P?  The density of TIP5P is different from the literature value of 
982. Thanks!


So your system is water + polymer?  In that case, you have an added variable - 
the interaction between the water and your polymer.  What you should probably be 
more concerned with the what the bulk density of the whole system is.  I have 
found that analyzing densities of portions of mixed systems does not give an 
accurate result.


-Justin



On Mon, May 18, 2009 at 3:33 PM, Mark Abraham > wrote:


Yanmei Song wrote:

Dear All:

I found when I changed the vdwtype from cut-off to shift, the
density of the system decrease significantly.All the other
setting in the mdp file remain unchanged. For example the
density of TIP5P water drops to 962. Does it mean I can not use
the shift for vdwtype?


Maybe not. Check out the paper in which TIP5P was parameterized. The
greater the extent to which your .mdp conditions differ from those
conditions, the greater the likelihood that the parameterization is
no longer valid.

Mark
___
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 gmx-users-requ...@gromacs.org
.
Can't post? Read http://www.gromacs.org/mailing_lists/users.php




--
Yanmei Song
Ph.D. Candidate
Department of Chemical Engineering
Arizona State University




___
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 gmx-users-requ...@gromacs.org.

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


--


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


___
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 gmx-users-requ...@gromacs.org.

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

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Yanmei Song 
Dear Mark:

Thanks for your answer. I have to use the same mdp file for both water and
another polymer molecules in my system, right?  Then if the shift function
works well for the polymer, I can not get the best result for TIP5P?  The
density of TIP5P is different from the literature value of 982. Thanks!

On Mon, May 18, 2009 at 3:33 PM, Mark Abraham wrote:

> Yanmei Song wrote:
>
>> Dear All:
>>
>> I found when I changed the vdwtype from cut-off to shift, the density of
>> the system decrease significantly.All the other setting in the mdp file
>> remain unchanged. For example the density of TIP5P water drops to 962. Does
>> it mean I can not use the shift for vdwtype?
>>
>
> Maybe not. Check out the paper in which TIP5P was parameterized. The
> greater the extent to which your .mdp conditions differ from those
> conditions, the greater the likelihood that the parameterization is no
> longer valid.
>
> Mark
> ___
> 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 gmx-users-requ...@gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>



-- 
Yanmei Song
Ph.D. Candidate
Department of Chemical Engineering
Arizona State University
___
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 gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [gmx-users] shift function for vdwtype and density

2009-05-18 Thread Mark Abraham 

Yanmei Song wrote:

Dear All:

I found when I changed the vdwtype from cut-off to shift, the density of 
the system decrease significantly.All the other setting in the mdp file 
remain unchanged. For example the density of TIP5P water drops to 962. 
Does it mean I can not use the shift for vdwtype?


Maybe not. Check out the paper in which TIP5P was parameterized. The 
greater the extent to which your .mdp conditions differ from those 
conditions, the greater the likelihood that the parameterization is no 
longer valid.


Mark
___
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 gmx-users-requ...@gromacs.org.

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

[gmx-users] shift function for vdwtype and density

2009-05-18 Thread Yanmei Song 
Dear All:

I found when I changed the vdwtype from cut-off to shift, the density of the
system decrease significantly.All the other setting in the mdp file remain
unchanged. For example the density of TIP5P water drops to 962. Does it mean
I can not use the shift for vdwtype?

-- 
Yanmei Song
Ph.D. Candidate
Department of Chemical Engineering
Arizona State University
___
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 gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/mailing_lists/users.php