Release of px and the Threading Analyst
rmiller at bsm.bioc.ucl.ac.uk
Tue Mar 28 20:18:28 EST 1995
I hereby announce the _official_ release of tan.tcl, the Threading
Analyst, and px, the Protein eXtensions to the Tcl/Tk scripting language.
This software is available by anonymous ftp from ftp.biochem.ucl.ac.uk
in the pub/px subdirectory.
The release includes full C source code for px, as well as (naturally)
Tcl/Tk/px code to support tan.tcl.
If you obtained an earlier release, _please_ (1) accept my apologies
for the many bugs you probably experienced and (2) get the new version.
Several new features have been added over the last few weeks, including
the ability to incorporate the results of separate analyses of the
threaded models. In addition, a _manual_ is also supplied.
What is the Threading Analyst ?
Protein threading is an approach to fold classification and prediction
based on modelling a given amino acid sequence as each of several folds
>From a pre-specified library. David Jones' Threader program that does
this, introduced in ``A new approach to protein fold recognition,''
(Nature, vol. 358 (1992), pp. 86-89) is available from our ftp server in
the pub/THREADER subdirectory (requires a user license -- free to
academics -- and an Irix 5.2 SGI or a Sun4). The Threading Analyst
(tan.tcl) is a GUI to sort and evaluate the Threader output for a
particular sequence, hopefully assisting the researcher in identifying
the best model(s) for her/his amino acid sequence or determining that the
fold is probably not in the current library. Given an appropriately
formatted file, tan.tcl will work with any data mapping an amino acid
sequence to a protein structure.
What is px ?
Px is a set of command extensions to John Ousterhout's Tcl/Tk scripting
language specifically for application in the protein structure
bioinformatics domain. Currently these commands primarily facilitate
reading, writing, and evaluating (via supplied potential functions)
protein structure and sequence data files, as well as providing support
the maintenance of citation data for encoded algorithms. Px is intended
to form the basis for a `toolbox' of encapsulated C functions for protein
structure analysis and prediction, allowing future applications to utilize
your old software in ways you haven't thought of yet. As you might imagine,
px is based on a set of C data structures that all commands can access.
Px commands are not dynamically loaded yet, but expect that improvement
soon. You are encouraged to add your own code to px, either submitting
it to me, posting it on the net, or keeping it for yourself (boo!). If
you have a need and ask nicely, I can probably supply you with my code to
build structures from dihedral angles (etc.) and/or the other structure
evaluation functions I'm working on.
px and associated Tcl/Tk software and source code is
copyright (c) 1995 Robert T. Miller except where noted in the
Development of px and the Threading Analyst has been supported by a grant
>From Glaxo Group Research, Greenford, Middlesex, United Kingdom; all
development effort was carried out in Janet Thornton's Biomolecular
Structure and Modelling Unit in the Department of Biochemistry and
Molecular Biology, University College, London, U.K.
Rob Miller, Ph.D.
Biomolecular Structure and Modelling Unit (BSM),
Department of Biochemistry and Molecular Biology,
University College / Gower Street / London WC1E 6BT.
Tel: 44 171387 7050 x2303
Fax: 44 171380 7193
Internet: rmiller at bsm.bioc.ucl.ac.uk
More information about the Bionews