home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!biosci!agate!stanford.edu!hsdndev!newsfeed.rice.edu!bcm!brent
- From: brent@kite.imgen.bcm.tmc.edu (Brent Wiese)
- Newsgroups: bionet.software
- Subject: Re: PIMA on SGI
- Keywords: protein sequence alignment
- Message-ID: <1k942lINN6ir@gazette.bcm.tmc.edu>
- Date: 28 Jan 93 17:10:45 GMT
- References: <1993Jan27.081224.21491@kakwa.ucs.ualberta.ca>
- Organization: Baylor College of Medicine, Houston, Tx
- Lines: 51
- NNTP-Posting-Host: kite.imgen.bcm.tmc.edu
- Originator: brent@kite.imgen.bcm.tmc.edu
-
-
- In article <1993Jan27.081224.21491@kakwa.ucs.ualberta.ca>, rndy@wimsey.mmid.ualberta.ca (Randy Read) writes:
- > I've downloaded Smith & Smith's PIMA (Pattern Induced Multiple Alignment)
- > package (MBCRR-Package/pima-1.22.tar.Z on mbcrr.harvard.edu) by ftp as
- > suggested in their paper. Attempts to install the package on either an SGI or
- > a NeXT lead to a number of compile-time warnings and fatal errors. I still
- > only speak pidgin C, so I'm hoping that someone with a greater facility with
- > the language has sorted this out.
- >
- > Has anyone out there succeeded in implementing this package on SGI hardware?
- > If so, is the revised version available, or is it just a matter of the right
- > compiler flags? I would greatly appreciate any information, either by e-mail
- > or posted to this newsgroup.
- >
- > Randy J. Read
- > rndy@wimsey.mmid.ualberta.ca
-
- Yes, your errors can be resolved by a compiler setting. Since the
- code is not ansi but rather K&R C you should tell the compiler this.
- The easiest way to do this would be to run ...
-
- setenv SGI_CC -cckr
-
- ... and then run the install.sh file. Additionally you will want to make
- as slight modification to the file make-cluster.sh . Line number 258 of
- make-cluster.sh should read ...
-
- ----------------------------------------------------------------
- awk '{print NR,$0}' tmp-cluster.$$ | extract-cluster-loci | awk '{print $2}' |
- ----------------------------------------------------------------
-
- ... instead of ...
-
- ----------------------------------------------------------------
- cat -n tmp-cluster.$$ | extract-cluster-loci | awk '{print $2}' |
- ----------------------------------------------------------------
-
- I don't have access to a NeXT system right now so I don't know how
- you can tell the C compiler on your NeXT system to expect K&R C, or if
- that is even possible. If you don't have anybody around there that knows
- how to tell the NeXT C compiler this, let me know and I will see what I can
- do. Please contact brent@bcm.tmc.edu or rsmith@bcm.tmc.edu if you are
- having any further difficulties.
-
- Thanks,
- Brent
- --
- Brent Wiese Phone: 713-798-6034
- Institute for Molecular Genetics Fax: 713-798-5386
- Baylor College of Medicine Internet: brent@bcm.tmc.edu
- Houston, TX 77030 USA
-