GIST: reference to bulk and column names

[halx]

Hi,

I am trying to find out which of the GIST columns are referenced to bulk water in order to compute the free energy. AMBER tutorial 25 suggests to compute the free energy from all the *-dens columns but the manual says that Eww-dens is not referenced. Interestingly, the column names in the tabular output file are Eww-dens and Eww-norm-unref. Do I need to assume that none of those two columns are referenced and I need to subtract the corresponding value from table 32.3 in the AMBER20 manual?

Many thanks/

[drroe]

Pinging @EricChen521 @acruzpr @jokr91 @gosldorf @cnnguyen @tkurtzman since they are the real GIST experts.

So as far as I can tell from the code, you are correct and the Eww- energies are not normalized w.r.t. bulk density. I believe the only quantities that make use of the bulk density are the solvent oxygen and hydrogen densities, the first-order translational entropies (dTStrans-), and second-order translational entropies (dTSsix-). Anyone pinged above please correct me if I'm wrong, or if I have missed something.

[EricChen521]

Hi,

Dan is correct that all Eww-dens, Esw-dens are not normalized. They represent the water-water / solute-water energy density in each voxel. The Eww-norm-unref or Eww-norm are the water-water energy or solute-water energy per water molecule in each voxel. In other words, E_norm=(E_dens * voxel_volume)/(ensembled water number of the voxel).

The referenced bulk density is only used for entropy calculation( TStran, TSori, TSsix)

Section 4b) of http://ambermd.org/tutorials/advanced/tutorial25/section4.htm may provide more details about the free energy calculation using GIST output.

[halx]

Many thanks. In the meantime, I have simulated a box of pure water and found delta G to be essentially zero for all waters (in a sub-box of arbitrary size) when calculated from all relevant *-dens columns and multiplying each by the voxel volume as suggested by the tutorial. The mean value for Eww-norm-unref for TIP3P water is essentially the same as the one reported in the table in the manual. The manual as it is now probably explains this all correctly. Nevertheless, I would suggest to add a line that simply explains how to get delta G (with respect to bulk) from the data available.

[DDGmichigan]

@EricChen521 Does GIST work only for water models or has it been implemented on organic solvents as well? I would be interested to learn if its made to tackle co-solvent hydration energies? Regards Debarati

[EricChen521]

Hi Debarati,

The GIST in CPPTRAJ only works for water models. However, the Liedl group recently introduced an extended GIST ( https://pubs.acs.org/doi/abs/10.1021/acs.jcim.0c00289) that can handle chloroform. Hope this helps.

Best, Eric

On Fri, Dec 25, 2020 at 9:53 AM DDGmichigan notifications@github.com wrote:

@EricChen521 https://github.com/EricChen521 Does GIST work only for water models or has it been implemented on organic solvents as well? I would be interested to learn if its made to tackle co-solvent hydration energies? Regards Debarati

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/Amber-MD/cpptraj/issues/849#issuecomment-751259414, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIH2DV3S2NRUQLMCN3YRO3LSWSRQJANCNFSM4R6X6MOQ .

[tkurtzman]

Hi DDG,

It's trivial to extend the GIST-cpptraj code to any rigid solvent. All you'd need to do is write code that calculates the quaternions to rotate one rigid molecule onto the other then feed that into the entropy calculations. The code we use in GIST right now for calculating the quaternions is specific to water molecules. Non-rigid solutes are more complicated though the sampling problems that come with them can likely be overcome with some reasonable approximations. These have both been on our to-do lists for some time though so far down in the queue that likely they won't be incorporated until needed for a specific scientific project.

If you'd like to extend the code (or even better generalize it to any rigid molecule) for your project, we'd be glad to help.

TK

On Fri, Dec 25, 2020 at 9:53 AM DDGmichigan notifications@github.com wrote:

@EricChen521 https://github.com/EricChen521 Does GIST work only for water models or has it been implemented on organic solvents as well? I would be interested to learn if its made to tackle co-solvent hydration energies? Regards Debarati

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/Amber-MD/cpptraj/issues/849#issuecomment-751259414, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGYKFQ24ZVQJ2HNEMXGH7TDSWSRQJANCNFSM4R6X6MOQ .

--

---

Tom Kurtzman, Ph.D. Associate Professor Department of Chemistry Lehman College, CUNY 250 Bedford Park Blvd. West Bronx, New York 10468 718-960-8832 http://www.lehman.edu/faculty/tkurtzman/ http://www.lehman.edu/faculty/tkurtzman/index.html

---