How to predict transmembrane protein topology?

[Raymond Mejia]

In article <4486ld$dpu at news.ust.hk>, bclong at uxmail.ust.hk (Terence) wrote:
>
>       Hello.
>
>       I often read in journals that a certain transmembrane topology is
>determined/predicted for a certain transmemnrane protein. How is this done?
>Is this done thru some commercial/public domain software? Or is it just by
>eyeballing how many hydrophobic segment there? 

A relatively naive approach is described in a paper, "Neural Net Models to
Characterize Integral Membrane Proteins - Preliminary Report" to appear in
Zeitschrift fuer Angewandte Mathematik und Mechanik (ZAAM) as part of the
Proceedings of ICIAM95.  A complete preprint is available via anonymous
ftp in file smb/pubs/Mejia_ICIAM95.tar.Z on fconvx.ncifcrf.gov
(129.43.51.13).  I welcome your comments and suggestions.

You may browse the the contents of the ftp archive via the Society for
Mathematical Biology Home Page at URL:
http://www.iam.ubc.ca/spider/spiros/smb/index.html. 

Regards,

Raymond Mejia
Laboratory of Kidney and Electrolyte Metabolism, NHLBI, and
Mathematical Research Branch, NIDDK
BSA/350
National Institutes of Health
Bethesda, MD  20892-2690
Tel: (301)496-9972                      Fax: (301)402-0535
E-mail: ray at helix.nih.gov               WWW: http://mrb.niddk.nih.gov/ray/