[Chimera-users] command 'buriedArea' - question

[Tom Goddard]

Hi Hari,

  This warning message happens because the Chimera code that computes 
solvent excluded surfaces does not always work.  The code is complex 
(~20000 lines of fortran) and we are working on replacing it with some 
more reliable code in the Chimera 1.4 release in a few months.

  Here's what the warning means and how to determine if there is a 
problem.  The surface calculation computes each connected part of the 
solvent excluded surface.  There might be disconnected patches of 
surface for two reasons.  The common one is that there are cavities on 
the interior of the protein, so a surface bubble appears on the inside 
of the protein and it is not connected to the exterior surface.  The 
second case is if the set of atoms is in 2 or more clumps that don't 
touch each other.  The message that calculating some surface components 
failed means that either that surfaces for some interior cavities or 
some extra exterior clumps of atoms could not be computed.  If you have 
disconnected clumps of atoms in either of your two atom sets that are 
involved in buried surface contacts and you get this warning then the 
reported areas will be wrong.  That is a somewhat unusual case.  If you 
have interior cavities, those generally are not part of the contacts and 
the reported areas will be correct.  The case to watch out for here is 
if the surface for the combined set of atoms has a cavity at the 
interface between the two atom sets, then that cavity won't be computed 
and the answer will be wrong.

  There are two things you can do.  You can try to get the surface 
computation to work using some tricks suggested in this mailing list message

    http://www.cgl.ucsf.edu/pipermail/chimera-users/2008-March/002417.html

or you can try to figure out whether your case is in fact giving correct 
areas in spite of the surface calculation problem.  For the latter you 
could make PDB files for each atom set, and for the combined atom sets, 
and use menu entry Actions / Surface / Show to create the three solvent 
excluded surfaces and see which ones fail to compute all components, and 
for those look at where the cavities are (at interface or not) and 
whether external atom clumps are present.

    Tom


-------- Original Message --------
Subject: command 'buriedArea' - question
From: Harinath
To: Tom Goddard <goddard at cgl.ucsf.edu>
Date: 8/12/09 9:57 AM
> Dear Tom,
>                      Here is the reply log when I used the 'measure 
> buried' command in a modeled complex. What am I missing in the pdb 
> file and how erroneous are the calculations?
>
> Please find the pdb file at the bottom to repeat the error.
> ______________________________________
> Calculation of some surface components failed.  Using only single 
> surface component.  This may give inaccurate areas if surfaces of 
> either set of atoms or the combined set are disconnected.
> Buried solvent accessible surface area = 1333.97 (A1 = 10302.4, A2 = 
> 7673.48, A12 = 15307.9)
> Buried solvent excluded surface area = 818.456 (A1 = 8044.45, A2 = 
> 6390.68, A12 = 12798.2)
> ______________________________________
>
> Sincerely,
> Hari
>
>
>
>
>
>
> On Aug 11, 2009, at 11:11 PM, Tom Goddard wrote:
>
>> Hi Hari,
>>
>> open 1a0m
>> measure buried :.A :.B
>>
>> This measures the buried area between chains A and B.  Here's the 
>> measure command documentation
>>
>>   http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/measure.html
>>
>> and some info on how to specify different sets of atoms
>>
>>   http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/atom_spec.html
>>
>> Tom
>>
>>
>>
>> -------- Original Message --------
>> Subject: [Chimera-users] command 'buriedArea'
>> From: Harinathachari Bahudhanapati
>> To: chimera-users at cgl.ucsf.edu
>> Date: 8/11/09 7:59 PM
>>> Hi all,
>>>         Could you explain the command 'measure buriedArea' ? I am 
>>> trying yesterday's daily build. Could you give me an example how to 
>>> use this command? I think I am mangling the command, somehow. Could 
>>> you help me?
>>>
>>> Thank you very much.
>>>
>>> Sincerely,
>>> Hari
>>>
>>
>