From: Ajith Rathnaweera Rajapaksha Mudalige (arajapak_at_purdue.edu)
Date: Sat Apr 21 2012 - 00:22:33 CDT
Hi Aron,
Thanks for trying to help.
I will check this one. I had not heard about this before.
Ajith
----- Original Message -----
From: "Aron Broom" <broomsday_at_gmail.com>
To: "Axel Kohlmeyer" <akohlmey_at_gmail.com>
Cc: "Ajith Rathnaweera Rajapaksha Mudalige" <arajapak_at_purdue.edu>, namd-l_at_ks.uiuc.edu
Sent: Saturday, April 21, 2012 12:46:53 AM
Subject: Re: namd-l: Simulation Using Both NAMD and VMD
As a kind of crazy side point here, doesn't the NAMD implicit solvation code utilize a Linear Combination of Partial Overlaps (LCPO) code to calculate not only SASA, but also uses that to generate a force in order to minimize SASA?
I may be totally off base on that.
~Aron
On Fri, Apr 20, 2012 at 9:28 PM, Axel Kohlmeyer < akohlmey_at_gmail.com > wrote:
On Fri, Apr 20, 2012 at 9:19 PM, Ajith Rathnaweera Rajapaksha Mudalige
< arajapak_at_purdue.edu > wrote:
> Hi Axel,
>
> measure sasa has other capabilities...
>
> It can compute the points on the solvent accessible surface and this can be done for each individual atoms.
> Forces are calculated using the coordinates of these points. So when the forces are in action surface changes and therefore forces are to be updated.
ok. i wasn't looking a the argument list of the code only at the return value.
> Everything I need can be done in NAMD, except computing measure sasa.
>
> Any suggestions?
if this is the case, passing the data from NAMD to VMD
and back is EVEN MORE STUPID, if that is at all possible.
why on earth do you insist on this insanity?
the SASA algorithm is not very difficult.
running two executables in step and
communicating data between them is,
not to mention the insane overhead of
formatting numbers into text, communicating
them and then parsing it and converting
to Tcl and back.
axel.
>
> Ajith
>
> ----- Original Message -----
> From: "Axel Kohlmeyer" < akohlmey_at_gmail.com >
> To: "Ajith Rathnaweera Rajapaksha Mudalige" < arajapak_at_purdue.edu >
> Cc: "Aron Broom" < broomsday_at_gmail.com >, namd-l_at_ks.uiuc.edu
> Sent: Friday, April 20, 2012 8:42:49 PM
> Subject: Re: namd-l: Simulation Using Both NAMD and VMD
>
> On Fri, Apr 20, 2012 at 7:14 PM, Ajith Rathnaweera Rajapaksha Mudalige
> < arajapak_at_purdue.edu > wrote:
>> Hi
>>
>> Thanks Alex, Aron,
>>
>> Yes, practically I will want to update the forces after every 500 time steps or so.
>>
>> I know this will slow down the simulation but it is an essential step to run the simulation correctly.
>>
>> The forces I am applying to my system is depending on Solvent Accessible Surface (SAS) of the molecule. SAS can be determined using "measure SASA" command in VMD. This is not directly supported in NAMD (please correct me if I am wrong). Therefore, I need to run the simulation in NAMD load the current phase in VMD and determine SAS at that point, determine the direction of the forces, get back to NAMD and run for the next phase.
>
> wait...
>
> there is one more thing that doesn't add up. measure sasa gives
> your a single number: the area. how does this compute into a
> directional force?
>
> axel.
>
>>
>> Is this possible?
>>
>> I will highly appreciate if anyone can give me a little concrete example, if anyone has done this type of a work before.
>>
>> Regards,
>>
>> Ajith
>>
>>
>>
>>
>> ----- Original Message -----
>> From: "Aron Broom" < broomsday_at_gmail.com >
>> To: "Axel Kohlmeyer" < akohlmey_at_gmail.com >
>> Cc: "Ajith Rathnaweera Rajapaksha Mudalige" < arajapak_at_purdue.edu >, namd-l_at_ks.uiuc.edu
>> Sent: Friday, April 20, 2012 6:30:14 PM
>> Subject: Re: namd-l: Simulation Using Both NAMD and VMD
>>
>> Is it possible to use the tcl forces capability of NAMD to run the script you want? As Axel points out, passing the information into VMD at every timestep would be crippling (unless your simulation takes an extremely long time per timestep already). If you only need to update the forces every 1000 timesteps or something, that might work.
>>
>> ~Aron
>>
>>
>> On Fri, Apr 20, 2012 at 6:10 PM, Axel Kohlmeyer < akohlmey_at_gmail.com > wrote:
>>
>>
>>
>> On Fri, Apr 20, 2012 at 6:05 PM, Ajith Rathnaweera Rajapaksha Mudalige
>> < arajapak_at_purdue.edu > wrote:
>>> Hi Everybody,
>>>
>>> Is it possible to write a script for a NAMD simulation to access VMD at each time step through Interactive Molecular Dynamics mode (or any other possible way).
>>
>> technically yes. in practice no, since that would be highly inefficient.
>> better to implement what you need into NAMD directly.
>>
>> axel.
>>
>>
>>
>>> In my simulation I need to execute some VMD commands at each time step and define a new force according to the results output from VMD.
>>>
>>> I highly appreciate if someone could shed some light in this direction.
>>
>>
>>
>>
>>> Regards,
>>>
>>> Ajith
>>>
>>
>>
>>
>> --
>> Dr. Axel Kohlmeyer
>> akohlmey_at_gmail.com http://goo.gl/1wk0
>>
>> College of Science and Technology
>> Temple University, Philadelphia PA, USA.
>>
>>
>>
>>
>> --
>> Aron Broom M.Sc
>> PhD Student
>> Department of Chemistry
>> University of Waterloo
>
>
>
> --
> Dr. Axel Kohlmeyer
> akohlmey_at_gmail.com http://goo.gl/1wk0
>
> College of Science and Technology
> Temple University, Philadelphia PA, USA.
-- Dr. Axel Kohlmeyer akohlmey_at_gmail.com http://goo.gl/1wk0 College of Science and Technology Temple University, Philadelphia PA, USA. -- Aron Broom M.Sc PhD Student Department of Chemistry University of Waterloo
This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:21:54 CST