VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Mon Feb 07 2005 - 19:48:40 CST

Edward,
  The restricted selection basically only allows atoms that are in
the restricted set to contribute to the calculated surface area.

I did the same test on my machine here on the PDB structure '1e79'
and I get results which are much closer, within a couple units of
surface area actually:

vmd > measure sasa 1.4 $protein -restrict $sidechain
100399.398438
vmd > measure sasa 1.4 $protein -restrict $backbone
32826.3398438

Sum of sidechain + backbone:
133225.7382818

vmd > measure sasa 1.4 $protein
133224.015625

"protein" minus sum of "sidechain" and "backbone":
1.7226568

I'm not sure why there's an 8% difference in your test case, but
here's something that may help clarify what you're getting.
The easiest thing to do in order to see what's being counted
and what isn't, is to use the -points option to store the generated
area sample points in a list, which can then be rendered using the
VMD draw commands:
  set sel [atomselect top all]
  set surf [atomselect top "sidechain"]
  set sasa [measure sasa 1.4 $sel -points pts -samples 50 -restrict $surf]
  puts "The SASA is $sasa"
  foreach pt $pts {
    draw point $pt radius 0.5
  }

That should make it much easier to visualize what is being included
and what is not.

  John

On Mon, Feb 07, 2005 at 01:47:34PM -0500, Edward Patrick Obrien wrote:
> Hi John,
> I've tried to figure out how to define this "restricted selection"
> correctly. I've one tentative conclusion, that I need you to confirm or
> reject:
>
> The restricted selection ("restrictedsel" according to documentation) MUST
> be a subset of the "selection", to get the SASA of the "restrictedsel".
>
> For example, to get the correct SASA of the sidechains (that is given
> there is also a backbone that excludes some area), I define and measure
> the sasa in the following manner:
>
>
> vmd > set sidechain [atomselect top sidechain]
> atomselect0
> vmd > set backbone [atomselect top backbone]
> atomselect1
> vmd > set protein [atomselect top protein]
> atomselect2
>
> vmd > measure sasa 1.4 $protein -restrict $sidechain
> 1253.48669434
> vmd > measure sasa 1.4 $protein -restrict $backbone
> 210.18057251
>
> Is this correct? Please let me know your thoughts.
>
> It is strange though, the unrestricted protein SASA is
>
> vmd > measure sasa 1.4 $protein
> 1583.38793945
>
> which still DOES NOT ADD UP from the components,
> that is:
>
> 1583 is not equal to 1253+210.
>
> Is this 8 percent difference acceptable?
>
> Thanks,
> Ed
>
>
>
>
>
>
> PS- below are the transcripts of my atempts:
>
> # Define the selections
> vmd > set sidechain [atomselect top sidechain]
> atomselect0
> vmd >
> vmd > set backbone [atomselect top backbone]
> atomselect1
> vmd > set protein [atomselect top protein]
> atomselect2
>
> # The unrestricted SASA
> vmd > measure sasa 1.4 $protein
> 1583.38793945
> vmd > measure sasa 1.4 $sidechain
> 1687.36560059
> vmd > measure sasa 1.4 $backbone
> 759.833068848
>
> # Incorrect, returns same SASA as unrestricted or 0
> vmd > measure sasa 1.4 $sidechain -restrict $sidechain
> 1687.36560059
> vmd > measure sasa 1.4 $sidechain -restrict $backbone
> 0.0
> vmd > measure sasa 1.4 $sidechain -restrict $protein
> 1687.36560059
> vmd > measure sasa 1.4 $backbone -restrict $sidechain
> 0.0
>
> # Try a new selection
> vmd > set bands [atomselect top "backbone and sidechain"]
> atomselect5
> vmd > measure sasa 1.4 $sidechain -restrict $bands
> 0.0
> vmd > measure sasa 1.4 $bands -restrict $sidechain
> 0.0
>
> # May be correct approach
> vmd > measure sasa 1.4 $protein -restrict $sidechain
> 1253.48669434
> vmd > measure sasa 1.4 $protein -restrict $backbone
> 210.18057251
>
>
> On Mon, 7 Feb 2005, Edward Patrick Obrien wrote:
>
> >
> >Hi John,
> > I'm confused about what the "selection" is in the restrict option.
> >I read the documentation and can't determine which of the following two is
> >correct if I want to get the sasa of the sidechains:
> >
> >atomselect sidechain [atomselect top sidechain]
> >atomselect backbone [atomselect top backbone]
> >
> >measure sasa 1.4 $sidechain -restrict $sidechain [1]
> >measure sasa 1.4 $sidechain -restrict $backbone [2]
> >
> >If I want to the sasa of the sidechains which restrict selection is
> >correct, [1] or [2]?
> >
> >Thanks,
> >Ed
> >
> >On Mon, 7 Feb 2005, John Stone wrote:
> >
> >>
> >>Ed,
> >> Since you're using b4, I think that the likely issue is that you
> >>need to add the restrict option to NOT count the surface area of
> >>atoms that border the parts you're excluding from the atom selection.
> >>The same would be true if you had internal cavities that you didn't want
> >>contributing to surface area.
> >>
> >> John Stone
> >> vmd_at_ks.uiuc.edu
> >>
> >>On Mon, Feb 07, 2005 at 12:49:18PM -0500, Edward Patrick Obrien wrote:
> >>>Hi John,
> >>> Thanks for the quick response.
> >>>The version is:
> >>> VMD for LINUX, version 1.8.3b4 (January 20, 2005)
> >>>
> >>>I've attached a graph of the time dependence of the sasa for
> >>>the protein, sidechains and backbone. I've checked several times
> >>>and I did not mislabel the files:)
> >>>
> >>>Ed
> >>>
> >>>PS- should I use the "restrict" option to correct this?
> >>>
> >>>On Mon, 7 Feb 2005, John Stone wrote:
> >>>
> >>>>
> >>>>Edward,
> >>>>Which exact beta version did you encounter this behavior with?
> >>>>We fixed a SASA calculation bug back in beta 3, so if you have a
> >>>>version older than that, that's the likely cause of the problem.
> >>>>Let me know which version you were testing with.
> >>>>
> >>>>Thanks,
> >>>>John Stone
> >>>>vmd_at_ks.uiuc.edu
> >>>>
> >>>>On Mon, Feb 07, 2005 at 12:26:16PM -0500, Edward Patrick Obrien wrote:
> >>>>>Hi All,
> >>>>> I downloaded the VMD v1.8.3 beta version to use the SASA command to
> >>>>>measure solvent accesible surface area.
> >>>>>
> >>>>> I can analyze my dcd's with it, BUT I am getting a physically
> >>>>>unreasonable result. The result is that the Surface area of the
> >>>>>sidechains
> >>>>>is larger then the Surface area of the protein as a whole!!!
> >>>>>
> >>>>> The SA,protein = SA,sidechains + SA,backbone, should be true. But
> >>>>>instead I find:
> >>>>> SA,protein < SA,sidechains + SA,backbone
> >>>>>
> >>>>>with the SA,sidechains > SA,protein.
> >>>>>
> >>>>> My initial guess is the sasa command is not taking into account the
> >>>>>excluded volume of the backbone when it measures the surface area of
> >>>>>the
> >>>>>sidechains, and vice-versa.
> >>>>>
> >>>>> Any suggestions about how to get self-consistent results?
> >>>>>
> >>>>>Thanks!
> >>>>>Ed
> >>>>>
> >>>>>PS- for srad I use 1.4, is this resonable?
> >>>>
> >>>>--
> >>>>NIH Resource for Macromolecular Modeling and Bioinformatics
> >>>>Beckman Institute for Advanced Science and Technology
> >>>>University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> >>>>Email: johns_at_ks.uiuc.edu Phone: 217-244-3349
> >>>>WWW: http://www.ks.uiuc.edu/~johns/ Fax: 217-244-6078
> >>>>
> >>
> >>
> >>
> >>--
> >>NIH Resource for Macromolecular Modeling and Bioinformatics
> >>Beckman Institute for Advanced Science and Technology
> >>University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> >>Email: johns_at_ks.uiuc.edu Phone: 217-244-3349
> >> WWW: http://www.ks.uiuc.edu/~johns/ Fax: 217-244-6078
> >>
> > 

-- 
NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
Email: johns_at_ks.uiuc.edu                 Phone: 217-244-3349              
  WWW: http://www.ks.uiuc.edu/~johns/      Fax: 217-244-6078