AMBER Archive (2009)

Subject: Re: [AMBER] The Right Neutralization Procedure and Ionic Strength Calculation ?

From: Marek Maly (marek.maly_at_ujep.cz)
Date: Fri Jul 17 2009 - 13:37:02 CDT


Dear Thomas,

thank you very much for your prompt and complex comments !

Just let me ensure that I undrestood well the part about "ionic strength".

I would like to focus just on the question of the NaCl ionic strength (not
overal one
which comprises all charges including that on my polyelectrolyte
macromolecules).

So let say that I have information that physiological IS of NaCl is 0.15 M
and
I would like to carry out my simulation under the same condition. So:

A)
I create and solvate my polyelectrolyte system.

B)
I add sufficient number of ions of one sign to perform
"net-neutralization"
( option #1 from my previous email).

C)
I add excess salt to reach concentration of 0.15 M of the NaCl molecules
without taking
in account "net-neutralization" counterions.

I thing that this approach could be in agreement with statement:

"I did computer experiment under physiological salt ionic strength
condition" although
the more precize statement is probably:

"I did computer experiment under BULK physiological salt ionic strength
condition"

But I think it is OK since that reported 0.15 M of NaCl is salt IS of the
bulk solution
(it means without my charged macromolecules or sufficiently far from them).

The rest topic is of course the reference volume which should be used for
the
IS calculation. I like the idea of using overal volume of water molecules
in given conditions
since this is in my opinion the true reference volume regarding to
definition of concentration
of some substance in water. If we use as the reference volume whole simul.
box we could obtain some
error since inside solvated macromolecules could exist some "forbidden"
volumes for water molecules and
also volume of that macromolecules could be NON-negligible.

But of course if we choose this way to calculate reference volume then we
have to solve problem
which is the volume of one water molecule. I would prefare to take volume
of one water molecule
regarding to used water model (for example TIP3P) although this in turn
requires one NPT simulation of pure
TIP3P box.

OK, one thing is the right interpretation/connections of relevant
parameters regarding to experimental ones but maybe would be
also a good idea to know precize definition/interpretation of relevant
MM-PBSA parameters ISTRNG (in PB) SALTCON (in GB).
Are they related only to concentration of excess salt ?

Thanks again for any additional comments, especially in case that I did
not undrestood
properly your IS interpretation !

Best wishes

   Marek

Dne Fri, 17 Jul 2009 18:50:15 +0200 Thomas Cheatham III <tec3_at_utah.edu>
napsal/-a:

>
>> My first question is related to neutralization procedure.
> ...
>> Let's assume that N1>N2 (overal charge is positive).
>>
>> Here I have two possibilities how to neutralize given system:
>>
>> #1
>> Add just n=N1-N2 Cl- counterions.
>
> "net-neutralization"
>
>> #2
>> Add N1 Cl- counterions and N2 Na+ counterions.
>
> "excess salt"
>
> The third possibility is no salt (which although unphysical due to the
> overall net-charge works in simulation since the Ewald forces do not
> diverge and then there is also this effective net-neutralizing plasma; in
> the past AMBER had two options whereby you could smear out the excess
> charge or use the implicit plasma).
>
> In 1998, with DNA, I did simulations on a ns time scale and saw little
> difference with each scheme. However, recent posts to the reflector
> pointed out some anomalies with RNA equilibration where it extended
> drastically if no salt was added... I am able to reproduce this and
> recommend you do either #1 or #2. Generally I personally try to match
> experiment so I try to find an added salt concentration close to
> experimental conditions (usually ~200 mM).
>
> The argument for not adding excess salt with polyanions like nucleic
> acids relates to sampling (since you would have to better sample the
> ionic distribution) and force field parameters (which were flawed and
> led to artificial crystallization)... However, as RNA can bind anions,
> it is probably wise to have Cl- around too. Therefore I would vote for
> #2.
>
>> My second question is related to ionic strength (IS).
>
> Ionic strength is complicated since you have the solute charges and also
> ion charges that contribute. The way I logically get around this is to
> specify that I add net-neutralizing salt to balance the overall charge
> and
> then add a certain concentration of excess salt (like 200 mM). The
> overall ionic strength should include all charges; however you could also
> report the ionic strength of the added salt.
>
>> the reference volume (for calculation of NaCl concentration) is the
>> overal
>> volume of water molecules as it is reported here:
>>
>> http://www.ks.uiuc.edu/Research/vmd/mailing_list/vmd-l/3943.html
>
> I have a perl script that I plug in the box size and molarity I want and
> it, based on volume (**not correcting for the volume of solute**), tells
> me how many ions to add. Remember also that by default LEaP produces
> water boxes that are slightly less dense than expected. I append the
> perl
> script below.
>
> --tec3
>
>
> molarity.perl
>
> #!/usr/sbin/perl
>
> $molar = $ARGV[0];
> $box_x = $ARGV[1];
> $box_y = $ARGV[2];
> $box_z = $ARGV[3];
> $box_a = $ARGV[4];
> $box_b = $ARGV[5];
> $box_g = $ARGV[6];
>
>
> if ($box_x == 0) {
> printf("Usage: molarity box-x box-y box-z alpha beta gamma\n");
> die;
> }
>
> if ($box_y == 0) {
> $box_y = $box_x;
> }
>
> if ($box_z == 0) {
> $box_z = $box_x;
> }
>
> if ($box_a == 0) {
> $box_a = 90.0;
> }
>
> if ($box_b == 0) {
> $box_b = $box_a;
> }
>
> if ($box_g == 0) {
> $box_g = $box_a;
> }
>
> $torad = 2 * 3.141592654 / 360.0;
>
> $rad_a = $box_a * $torad;
> $rad_b = $box_b * $torad;
> $rad_g = $box_g * $torad;
>
> $angles = 1 - cos($rad_a)*cos($rad_a) - cos($rad_b)*cos($rad_b) -
> cos($rad_g)*cos($rad_g);
> $angles += 2 * cos($rad_a)*cos($rad_b)*cos($rad_g);
> $angles = sqrt($angles);
>
>
> $volume = $box_x * $box_y * $box_z * $angles;
>
>
> $molecules = 6.022 * $volume * $molar / 10000;
>
> printf(" MOLARITY = %8.3f\n", $molar);
> printf(" Box size = %8.3f %8.3f %8.3f %8.3f %8.3f %8.3f\n", $box_x,
> $box_y, $box_z, $box_a, $box_b, $box_g);
> printf(" Volume = %8.3f\n", $volume);
> printf("\n %8.3f molecules are necessary to make a molarity of %6.2f
> M\n\n", $molecules, $molar);
>
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER_at_ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
> __________ Informace od NOD32 4051 (20090504) __________
>
> Tato zprava byla proverena antivirovym systemem NOD32.
> http://www.nod32.cz
>
>

-- 
Tato zpráva byla vytvořena převratným poštovním klientem Opery:  
http://www.opera.com/mail/

_______________________________________________ AMBER mailing list AMBER_at_ambermd.org http://lists.ambermd.org/mailman/listinfo/amber