Michel Seigneuret (seigneur at lpbc.jussieu.fr) wrote:
: Dear Collegues,
: I would like to use Grasp in order to display the local hydrophobicity
: on the molecular surface of a non-charged peptide. For this I would
: prefer to use atom-based rather than residue-based hydrophobicity
: values (e.g. as is apparently done for QSAR analysis of small
: molecules).Could somebody indicate me if such a method to calculate
: atom-based hydrophobicity parameters for proteins does exist, with
: reference to some publications or to an available program ?
: Thanks in advance.
One way that I've looked at this in Grasp is to use Grasp's excellent
treatment of electrostatics and twist it into displaying hydrophobicity.
The way I did this is to take Kyte-Doolittle hydrophobicity measures
and equate the hydrophobicity to charge. Basically, make a new charging
file that uses Kyte-Doolittle hydrophobicity spread over the atoms in
each residue in some semi-intelligent way.
Yeah, I know this might be considered completely bogus, but the results
are not too crazy and look GREAT on the screen. We've in fact used this
new hydrophobicity measure to discover some previously hidden features
about proteins we've been looking at, including dimerization interfaces,
active site identification and lots more.
For more information and my charging file, write to me.
Steven Ness
sness at byron.biochem.ubc.ca
Systems Assistant MP
Strynadka Lab
Department of Biochemistry
University of British Columbia
Canada
