IUBio

PairWise and SearchWise 1.5 beta release

Ewan Birney birney at molbiol.ox.ac.uk
Sun Feb 4 13:14:21 EST 1996


**** PairWise and SearchWise 1.5 Release   ****
         ***                                       ***


   This is to announce the beta release of PairWise and SearchWise.

        The programs PairWise and SearchWise are based around an
algorithm which compares a Protein Profile to a DNA sequence,
*allowing* for frameshifts and introns in the DNA sequence. As a
protein profile can be trivially made from a protein sequence, both
PairWise and SearchWise can use a protein sequence as a query.

        This is the first proper beta release for PairWise and
SearchWise.  Although there are a variety of bugs to be worked out,
and many aspects to improve, the core of PairWise and SearchWise is
probably robust enough to be used on additional sites with not too
much work :)

        PairWise also comes with Profileweight code in-built, which
has many options for the optimisation of profile production.

        More information on PairWise and SearchWise can be found at

        http://www.ocms.ox.ac.uk/~birney/wise/topwise.html

which includes a WWW based server for you to try out PairWise queries
at

        http://www.ocms.ox.ac.uk/~birney/wise/hwise.html

and binaries can be found at the ftp site

        nmrz.ocms.ox.ac.uk in /pub/wise but I would read the hypertext
help first!

        See the README file on the ftp site or any of the source code
files for copyright (it is also on the WWW site). It is free for both
academic and commerical users, but commerical users must register
first.
        
        As a teaser, in one SearchWise run we discovered around 40% of
the ESTs with a PH domain to contain frameshifts, almost certainly due
to seqeuncing errors.  These sequences were unlikely to be seen
without SearchWise - although the GCG Framesearch program can cope
with frame shifts it cannot use the more sensitive profile methods
which can find distantly related sequences.

        PairWise and SearchWise are designed to work either as
stand-alone programs or in the GCG sequence analysis
enviroment. PairWise can be linked to GCG8 at compile time.
SearchWise can read GCG databases (ascii or binary), fasta databases
or EMBL databases.

        PairWise and SearchWise are written in ANSI C and developped
concurrently on SunOS, IRIX and AXP VMS platforms. Digital UNIX seems
to require no changes. It is likely that programs are easily ported to
any platform with an ANSI C compiler.

        The front ends (PairWise and SearchWise) have been tweaked
with to make sure that novice users can easily grasp what to do and 
not to make (too many) mistakes early on. However the user is given
complete control over as many aspects as possible and can reset
virtually all options through a run time configuration system.

        There are a few bugs in: In these cases they are documented if
at all possible in both output and log files. I respond quite quickly
to bugs: any suggestions are also very wellcome. 


        I hope you find the package as useful as we have over the years.


Author/Programmers:

        Ewan Birney (Oxford)
        Julie Thompson (EMBL)   

Author/Developper/Guru and Lead Bug tester ;)

        Toby Gibson (EMBL)



More information about the Info-gcg mailing list

Send comments to us at biosci-help [At] net.bio.net