From owner-metabolic-reg@net.bio.net Tue Oct 05 08:53:00 1999 Path: biosci!newshost.lanl.gov!logbridge.uoregon.edu!newsfeed.direct.ca!tank.news.pipex.net!pipex!server1.netnews.ja.net!pegasus.csx.cam.ac.uk!hgmp.mrc.ac.uk!biosci From: JHSH@LAND.SUN.AC.ZA ("Jannie Hofmeyr") Newsgroups: bionet.metabolic-reg Subject: Re: how do I define the l mtarix from MCA in maple? Date: 5 Oct 1999 10:45:42 +0100 Organization: Stellenbosch University Lines: 36 Message-ID: References: <939110454.6941.0.nnrp-08.c2de3927@news.demon.co.uk> NNTP-Posting-Host: mercury.hgmp.mrc.ac.uk Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT X-Trace: niobium.hgmp.mrc.ac.uk 939116743 15494 193.62.192.80 (5 Oct 1999 09:45:43 GMT) X-Complaints-To: news@hgmp.mrc.ac.uk NNTP-Posting-Date: 5 Oct 1999 09:45:43 GMT X-Received: from mail.sun.ac.za (mail.sun.ac.za [146.232.128.1]) by hgmp.mrc.ac.uk (8.9.3/8.9.3) with ESMTP id KAA11005 for ; Tue, 5 Oct 1999 10:45:36 +0100 (BST) X-Received: from land.sun.ac.za ([146.232.216.2]) by mail.sun.ac.za with esmtp (Exim 2.10 #1) id 11YR9m-0003iK-00 for btk-mca@hgmp.mrc.ac.uk; Tue, 5 Oct 1999 11:45:14 +0200 X-Received: from SUN_LAND/SpoolDir by land.sun.ac.za (Mercury 1.44); 5 Oct 99 11:44:12 +0200 X-Received: from SpoolDir by SUN_LAND (Mercury 1.44); 5 Oct 99 11:43:46 +0200 X-To: btk-mca@hgmp.mrc.ac.uk X-Reply-to: jhsh@mail.sun.ac.za X-Priority: normal X-In-reply-to: <939110454.6941.0.nnrp-08.c2de3927@news.demon.co.uk> X-Mailer: Pegasus Mail for Win32 (v3.12) Herbert wrote > I'm sure there is more than one way to skin the cat here (apologies to cat > lovers but I'd be interested to learn of other cat skinning approaches with > regard to the link matrix) but one way to compute the L matrix is to use the > following algorithm: > > Algorithm: L = tr (echelon (tr (m))) > > where tr is the transpose, echelon computes the fully *reduced* row echelon and > m is the stoichiometry matrix. I'm not sure what the syntax would be in matlab. > Also I think there is a way of using svd which is probably the recommended way > but I haven't explored this avenue. > > Not that the reduction of the stoichiometry matrix to a fully reduced > echelon may involve row exchanges which means you *must* carry out the same > exchanges in the stoichiometry matrix. Herbert is of course quite correct. I think Pedro implements the same algorithm in Gepasi (do you Pedro?). What about Stefan Schuster's algorithm? - being based on convex algebra it has the nice property of providing 'sensible' conservation equations. Has anyone ever tried SVD? I played around with it a bit some time ago, but did not pursue the matter to a point where it was an alternative. Jannie Prof. Jannie Hofmeyr Tel: +27 21 808 3039 Dept. of Biochemistry Fax: +27 21 808 3022 University of Stellenbosch e-mail: jhsh@maties.sun.ac.za Private Bag X1 Matieland 7602 Stellenbosch South Africa --- .