Hi all,
here's the original post:
At 10:55 AM 6/19/2007 +0200, you wrote:
Hi all,
a few days ago I sent a post in which I was asking if anybody knew a
program to automatically define the hydrophobic core of a protein,
given the pdb.
Unfortunately I got no answers, and indeed a more thorough googling
around revealed that such a program might not exist.
So it seems I have to define my hydrophobic core residues by hand...
So now my question would be: how to define the hydrophobic core residues?
I would tend to say that those that bury more than ## % (say 70%,
80% ??) of their otherwise solvent accessible surface area could be
defined as such, but how can I get such a per residue percentage?
(NB: this is not the asa buried upon interaction, so I don't know
how to get the asa of the "free" amino acid)
Alternatively, are there other simple and defined rules to state
which are the hydrophobic core residues?
Any help appreciated,
thanks in advance,
ciao
s
I received several answers, and wish to thank all of those who took
their time to help me out on this issue.
I finally used naccess, which is very nice and gives you a percentage
of buried accessible surface area with respect to the asa of the
amino acid in an extended conformation (calculated in an Ala-X-Ala peptide).
I found that a nice definition of the hydrophobic core amino acid
corresponds to those that bury more than 90% of their asa.
Here's a quick summary of the most relevant replies.
thanks again,
ciao
s
[EMAIL PROTECTED]:
>free amino acid: ASA gly/ala-X-gly/ala, and program: naccess. mind you
>residues close to surface residues mihgt easily have 70% buried... i would
>be more conservative. check first with 95%.
>
>hth,
>tommi
mmerckel <[EMAIL PROTECTED]>:
>Hi Sebastiano,
>
>You might try NACCESS for obtaining atom and residue exposed surface areas.
>It returns absolute area and percent area relative to that residue in the
>middle position of a tripeptide. You should also look at Merk Gerstein's
>CODE_MBG programs for some information on buried residue calculations.
>Also look for DPX, residue depth calculation, there is a server and
>free standing C code.
>
>I also have an old reference to hydrophobic core calculations,
>perhaps PEDS or Protein Science, but I haven't (re)found it yet.
>Someone also sent me code for this, if I can find it I'll forward.
Liz Potterton <[EMAIL PROTECTED]>:
>Sebastiano,
>
>I can suggests one approach to finding core residues..
>
>CCP4mg will calculate the total asa buried per residue and residues can be
>selected by the criteria of their buried area. One way to define
the protein
>core would be to select residues with, say, <5.0A*2 asa. You can easily see
>what has been selected in the molecular graphics and try varying
the criteria
>if necessary.
>If you need to do this for many structures we could write a script.
>
>Liz
"Judith Murray-Rust" <[EMAIL PROTECTED]>:
>Sebastiano
>
>Naccess will calculate acessibility relative to the same residue in an
>extended peptide, which sounds pretty much what you want. It's not very
>automatic, however. YOu have to install it
>http://wolf.bms.umist.ac.uk/naccess/
>
>and either human-read or parse the output.
>
>J
"Nadir T. Mrabet" <[EMAIL PROTECTED]>:
>Hi,
>
>There exists a formal analytical way in doing this using the
"survol" command in BRUGEL package.
>In short, this command define the accessible surface (external and
cavities) to the probe you choose and creates
>separate masks (ensembles) of all atoms/residues that define these surfaces.
>The way to go, then, would be to create a collection mask of all
accessible atoms/residues and subtract this
>from your protein mask to be left with the mask that contains the
core residues.
>You could also use a cutoff of 5-10% ASA max (there are different
ways of calculating these !).
>On the other hand, taking away even residues that display very
little ASA (< 5%) would certainly leave you with
>genuine core residues.
>Hope this helps.
>
>Greetings,
>
>Pr. Nadir T. Mrabet
"Shabir Najmudin" <[EMAIL PROTECTED]>:
>Dear Sebastian,
>
>Have you looked at Simon Hubbard's NACCESS program? I think this is
what yo uare looking for. It does analysis per aa residue if you want.
>
>People at UCL-Biochemistry department - particularly Chrisitne
Orengo and Janet Thornton's group have done a vast amount analyses on
protein structures and have a lot of analysis >tools on their website.
>
>Hope this helps
>
>Cheers
>
>Shabir
Juergen Bosch <[EMAIL PROTECTED]>:
>Hi Sebastiano,
>
>you can misuse the Maestro suite from Schroedinger to define your
hydrophobic core. But I guess it would be an overkill to explain this
here. You have to fool the >program by indicating a faked ligand in
the region which you are looking for the core, then when you define
the constraints you can click on the hydrophobic >constraints tab and
it will show you the regions it things are hydrophobic in the closer
region where your faked ligand lies. Or you can simply look at your
PDB file by >coot and see the core yourself. That's how I do it in general.
>
>If you are a more graphics oriented guy, check out Pymol color by
atom, then look at those regions which are full of grey for your carbons.
>
>Juergen
Charlie Bond <[EMAIL PROTECTED]>:
>Hi Sebastiano,
>
>I think the CCP4 program AREAIMOL will help you with this (I'm on the
>road and can't check). I think if you run areaimol on your protein it
>will list each residue with its solvent accessible surfaces. You
>should be able to select those with, for example <5% accessible
>surface and define them as buried.
>
>Cheers,
>Charlie
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.476 / Virus Database: 269.9.6/865 - Release Date: 6/24/2007 8:33 AM
