From pmr at ebi.ac.uk Thu Mar 1 04:47:19 2007 From: pmr at ebi.ac.uk (Peter Rice) Date: Thu, 01 Mar 2007 09:47:19 +0000 Subject: [EMBOSS] question about translation start stie In-Reply-To: <4537.130.127.150.224.1172706465.squirrel@wm.clemson.edu> References: <4537.130.127.150.224.1172706465.squirrel@wm.clemson.edu> Message-ID: <45E6A127.6000403@ebi.ac.uk> Dear Fang, > Does anyone know if EMBOSS could give us the translation start site and > translation start site ? Thanks! > Looking forward to your reply. Can you give an example of what you mean? Start position and first codon perhaps? using the feature table, or from finding open reading frames? regards, Peter From mathog at caltech.edu Mon Mar 5 17:18:17 2007 From: mathog at caltech.edu (David Mathog) Date: Mon, 05 Mar 2007 14:18:17 -0800 Subject: [EMBOSS] Emboss 4.0.0 on Solaris 9, EMNU and MSE issues Message-ID: I'm having some problems getting EMNU and MSE to build on a Solaris 9 (Sparc) machine. Pretty sure in both cases it has to do with an ncurses versus curses confusion somewhere inside the configure script. After building EMBOSS 4.0.0 I was able to build all of the Embassy packages other than EMNU and MSE with just: ./configure make make install EMNU blew up at the make with: gcc -O2 -o .libs/emnu emnu.o ../../../nucleus/.libs/libnucleus.so ../../../ajax/.libs/libajax.so ../../../plplot/.libs/libplplot.so -lmenu -lform -L/usr/lib -lnsl -lsocket /usr/local/lib/libgd.so -L/usr/local/lib -L/usr/local/BerkeleyDB.4.2/lib -L/usr/local/ssl/lib -L/usr/openwin/lib -lXpm -lX11 /usr/local/lib/libjpeg.so /usr/local/lib/libfontconfig.so -L/usr/openwin/llib /usr/local/lib/libexpat.so /usr/local/lib/libfreetype.so -lpng12 -lpng -lz -lm /usr/local/lib/libiconv.so -lncurses -R/usr/local/lib -R/usr/lib Undefined first referenced symbol in file box32 emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) initscr32 emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) w32attron emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) w32attroff emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) acs32map emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) getbegx emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) getbegy emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) getcurx emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) getcury emnu.o (symbol belongs to implicit dependency /usr/lib/libcurses.so.1) Those are curses functions. To build EMNU it was only necessary to do cd source nedit Makefile (change -lncurses -> -lcurses) cd .. make make install and emnu worked. It would start, menu navigation was possible, and so forth. Trying the same thing with MSE allowed it to compile and link, but it wouldn't run. Or rather, it would start, but the screen would lock up. Here is where the build went wrong with the Makefile as originally created by ./configure: gcc -O2 -o .libs/mse mse.o ../ckit/.libs/libckit.so -L/usr/local/lib -L/usr/local/BerkeleyDB.4.2/lib -L/usr/local/ssl/lib -L/usr/openwin/lib -L/usr/lib -L/usr/openwin/llib ../../../nucleus/.libs/libnucleus.so ../../../ajax/.libs/libajax.so ../../../plplot/.libs/libplplot.so -lnsl -lsocket /usr/local/lib/libgd.so -lXpm -lX11 /usr/local/lib/libjpeg.so /usr/local/lib/libfontconfig.so /usr/local/lib/libexpat.so /usr/local/lib/libfreetype.so -lpng12 -lpng -lz -lm /usr/local/lib/libiconv.so -lncurses -R/usr/local/lib -R/usr/lib Undefined first referenced symbol in file initscr32 mse.o w32addch mse.o w32attrset mse.o The solaris 9 curses pieces are here: /usr/lib/libcurses.so /usr/include/curses.h and the ncurses pieces are here: /usr/local/lib/libncurses.so /usr/local/include/ncurses/curses.h Anybody else seen this? What needs to be done to get mse to build? Thanks, David Mathog mathog at caltech.edu Manager, Sequence Analysis Facility, Biology Division, Caltech From mathog at caltech.edu Tue Mar 6 14:50:30 2007 From: mathog at caltech.edu (David Mathog) Date: Tue, 06 Mar 2007 11:50:30 -0800 Subject: [EMBOSS] Emboss 4.0.0 on Solaris 9, EMNU and MSE issues Message-ID: Alan Bleasby wrote: > Your email suggests that you didn't configure mse using: > > ./configure --with-curses=/usr/local [other options] > > It is worth trying that if only to eliminate it as a possibility. That fixed both EMNU and MSE - they both built without problems and ran ok (where "ok" means the curses stuff worked in very limited testing.) Thanks, David Mathog mathog at caltech.edu Manager, Sequence Analysis Facility, Biology Division, Caltech From ajb at ebi.ac.uk Wed Mar 7 13:05:21 2007 From: ajb at ebi.ac.uk (ajb at ebi.ac.uk) Date: Wed, 7 Mar 2007 18:05:21 -0000 (GMT) Subject: [EMBOSS] EMBOSS 4.1.0 released Message-ID: <52224.81.98.244.247.1173290721.squirrel@webmail.ebi.ac.uk> EMBOSS 4.1.0 is available for download from: ftp://emboss.open-bio.org/pub/EMBOSS/ It contains: a) Two new applications, edialign and wordfinder b) Improvements to other applications e.g. coderet can now write non-coding regions to a file. c) Bugfixes, including all those supplied for the previous release. See the attached ChangeLog file for full details. Please remember that you should not install a new release over the top of an existing one. Also remember that the associated EMBASSY packages (e.g. MEME/PHYLIP/etc) are specific to a given release i.e. download them again if you wish to install them. A version for Microsoft Windows is also available at: ftp://emboss.open-bio.org/pub/EMBOSS/windows/ Alan -------------- next part -------------- A non-text attachment was scrubbed... Name: ChangeLog Type: application/octet-stream Size: 17537 bytes Desc: not available Url : http://lists.open-bio.org/pipermail/emboss/attachments/20070307/baab0d70/attachment-0001.obj From golharam at umdnj.edu Wed Mar 7 18:18:46 2007 From: golharam at umdnj.edu (Ryan Golhar) Date: Wed, 07 Mar 2007 18:18:46 -0500 Subject: [EMBOSS] EMBOSS 4.1.0 released In-Reply-To: <52224.81.98.244.247.1173290721.squirrel@webmail.ebi.ac.uk> Message-ID: <007301c7610e$f62ab960$2d01a8c0@PICO> There are three new files installed called tp400_dna, tp400_prot, and tp400_trans. What are these for? > -----Original Message----- > From: emboss-bounces at lists.open-bio.org > [mailto:emboss-bounces at lists.open-bio.org] On Behalf Of ajb at ebi.ac.uk > Sent: Wednesday, March 07, 2007 1:05 PM > To: emboss at emboss.open-bio.org > Subject: [EMBOSS] EMBOSS 4.1.0 released > > > EMBOSS 4.1.0 is available for download from: > ftp://emboss.open-bio.org/pub/EMBOSS/ > > It contains: > > a) Two new applications, edialign and wordfinder > b) Improvements to other applications e.g. coderet can now write > non-coding regions to a file. > c) Bugfixes, including all those supplied for the previous release. > > See the attached ChangeLog file for full details. > Please remember that you should not install a new release > over the top of an existing one. Also remember that the > associated EMBASSY packages (e.g. MEME/PHYLIP/etc) are > specific to a given release i.e. download them again if you > wish to install them. > > A version for Microsoft Windows is also available at: > ftp://emboss.open-bio.org/pub/EMBOSS/windows/ > > Alan > From ajb at ebi.ac.uk Wed Mar 7 19:34:48 2007 From: ajb at ebi.ac.uk (ajb at ebi.ac.uk) Date: Thu, 8 Mar 2007 00:34:48 -0000 (GMT) Subject: [EMBOSS] EMBOSS 4.1.0 released In-Reply-To: <007301c7610e$f62ab960$2d01a8c0@PICO> References: <52224.81.98.244.247.1173290721.squirrel@webmail.ebi.ac.uk> <007301c7610e$f62ab960$2d01a8c0@PICO> Message-ID: <51234.81.98.244.247.1173314088.squirrel@webmail.ebi.ac.uk> > There are three new files installed called tp400_dna, tp400_prot, and > tp400_trans. What are these for? They are data files for edialign. Alan From golharam at umdnj.edu Wed Mar 7 19:20:45 2007 From: golharam at umdnj.edu (Ryan Golhar) Date: Wed, 07 Mar 2007 19:20:45 -0500 Subject: [EMBOSS] EMBOSS and EMBASSY RPMs available Message-ID: <007b01c76117$9e1dbe80$2d01a8c0@PICO> The EMBOSS and EMBASSY RPMs have been updated for EMBOSS 4.1.0 All RPMs are available for download from http://informatics.umdnj.edu From golharam at umdnj.edu Wed Mar 7 20:33:30 2007 From: golharam at umdnj.edu (Ryan Golhar) Date: Wed, 07 Mar 2007 20:33:30 -0500 Subject: [EMBOSS] EMBOSS and EMBASSY RPMs available In-Reply-To: <007b01c76117$9e1dbe80$2d01a8c0@PICO> Message-ID: <008b01c76121$c7b9dc10$2d01a8c0@PICO> I apologize, a more direct URL is: http://informatics.umdnj.edu/BioRPMs > -----Original Message----- > From: emboss-bounces at lists.open-bio.org > [mailto:emboss-bounces at lists.open-bio.org] On Behalf Of Ryan Golhar > Sent: Wednesday, March 07, 2007 7:21 PM > To: emboss at emboss.open-bio.org > Subject: [EMBOSS] EMBOSS and EMBASSY RPMs available > > > The EMBOSS and EMBASSY RPMs have been updated for EMBOSS 4.1.0 > > All RPMs are available for download from http://informatics.umdnj.edu > > > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From fernan at iib.unsam.edu.ar Thu Mar 8 07:50:10 2007 From: fernan at iib.unsam.edu.ar (Fernan Aguero) Date: Thu, 8 Mar 2007 09:50:10 -0300 Subject: [EMBOSS] EMBOSS 4.1.0 released In-Reply-To: <52224.81.98.244.247.1173290721.squirrel@webmail.ebi.ac.uk> References: <52224.81.98.244.247.1173290721.squirrel@webmail.ebi.ac.uk> Message-ID: <20070308125010.GA32117@iib.unsam.edu.ar> +----[ ajb at ebi.ac.uk (07.Mar.2007 15:32): | | EMBOSS 4.1.0 is available for download from: | ftp://emboss.open-bio.org/pub/EMBOSS/ [snipped] | Also remember that the associated EMBASSY packages | (e.g. MEME/PHYLIP/etc) are specific to a given release i.e. download them | again if you wish to install them. | +----] I am updating the emboss FreeBSD port, and noticed that the embassy tarballs that were named as *NEW are not 'new' anymore (i.e. they were renamed back to the previous name, without the suffix). MEMENEW => MEME PHYLIPNEW => PHYLIP However, only the gzipped tarball file was renamed, the contents of the tarball extract to a directory containing the previous name: gama# tar tvzf PHYLIP-3.6b.tar.gz | less drwxrwxrwx 0 root root 0 Mar 6 08:07 PHYLIPNEW-3.6b/ drwxrwxrwx 0 root root 0 Mar 6 08:07 PHYLIPNEW-3.6b/m4/ -rw-r--r-- 0 root root 4626 Jan 31 09:48 PHYLIPNEW-3.6b/m4/java.m4 ... I can work around this, but certainly keeping the same distfile name in all places is cleaner and easier to work with :) Fernan From golharam at umdnj.edu Thu Mar 8 08:51:08 2007 From: golharam at umdnj.edu (Ryan Golhar) Date: Thu, 08 Mar 2007 08:51:08 -0500 Subject: [EMBOSS] EMBOSS 4.1.0 released In-Reply-To: <20070308125010.GA32117@iib.unsam.edu.ar> Message-ID: <009601c76188$d4b7f2a0$2d01a8c0@PICO> I agree. I was also expecting the version number on the tarballs to change as well. At the moment, there is no way to tell they were updated... > -----Original Message----- > From: emboss-bounces at lists.open-bio.org > [mailto:emboss-bounces at lists.open-bio.org] On Behalf Of Fernan Aguero > Sent: Thursday, March 08, 2007 7:50 AM > To: ajb at ebi.ac.uk > Cc: emboss at emboss.open-bio.org > Subject: Re: [EMBOSS] EMBOSS 4.1.0 released > > > +----[ ajb at ebi.ac.uk (07.Mar.2007 15:32): > | > | EMBOSS 4.1.0 is available for download from: > | ftp://emboss.open-bio.org/pub/EMBOSS/ > > [snipped] > > | Also remember that the associated EMBASSY packages > | (e.g. MEME/PHYLIP/etc) are specific to a given release i.e. > download > | them again if you wish to install them. > | > +----] > > I am updating the emboss FreeBSD port, and noticed that the > embassy tarballs that were named as *NEW are not 'new' > anymore (i.e. they were renamed back to the previous name, > without the suffix). > > MEMENEW => MEME > PHYLIPNEW => PHYLIP > > However, only the gzipped tarball file was renamed, the > contents of the tarball extract to a directory containing the > previous name: > > gama# tar tvzf PHYLIP-3.6b.tar.gz | less > drwxrwxrwx 0 root root 0 Mar 6 08:07 PHYLIPNEW-3.6b/ > drwxrwxrwx 0 root root 0 Mar 6 08:07 PHYLIPNEW-3.6b/m4/ > -rw-r--r-- 0 root root 4626 Jan 31 09:48 > PHYLIPNEW-3.6b/m4/java.m4 > ... > > I can work around this, but certainly keeping the same > distfile name in all places is cleaner and easier to work > with :) > > Fernan > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From pmr at ebi.ac.uk Thu Mar 8 09:43:54 2007 From: pmr at ebi.ac.uk (Peter Rice) Date: Thu, 08 Mar 2007 14:43:54 +0000 Subject: [EMBOSS] EMBOSS 4.1.0 released In-Reply-To: <009601c76188$d4b7f2a0$2d01a8c0@PICO> References: <009601c76188$d4b7f2a0$2d01a8c0@PICO> Message-ID: <45F0212A.4070204@ebi.ac.uk> Ryan Golhar wrote: > I agree. I was also expecting the version number on the tarballs to change > as well. At the moment, there is no way to tell they were updated... The embassy changes are all minor. We like to use the version numbers of the original code so it is a little difficult to merge in the EMBOSS version ... without making up a very long version number. Does anyone have strong preferences? Peter From golharam at umdnj.edu Thu Mar 8 09:59:34 2007 From: golharam at umdnj.edu (Ryan Golhar) Date: Thu, 08 Mar 2007 09:59:34 -0500 Subject: [EMBOSS] EMBOSS 4.1.0 released In-Reply-To: <45F0212A.4070204@ebi.ac.uk> Message-ID: <00b301c76192$638cf490$2d01a8c0@PICO> Maintaining the version number from the original code is important. I guess this is the nice thing about RPMs - it adds a release number after the version number for changes to the RPM which is what I've been using. What about adding a field to the version number to signify EMBOSS version relationships? > -----Original Message----- > From: Peter Rice [mailto:pmr at ebi.ac.uk] > Sent: Thursday, March 08, 2007 9:44 AM > To: golharam at umdnj.edu > Cc: 'Fernan Aguero'; emboss at emboss.open-bio.org > Subject: Re: [EMBOSS] EMBOSS 4.1.0 released > > > Ryan Golhar wrote: > > I agree. I was also expecting the version number on the > tarballs to > > change as well. At the moment, there is no way to tell they were > > updated... > > The embassy changes are all minor. We like to use the version > numbers of the > original code so it is a little difficult to merge in the > EMBOSS version ... > without making up a very long version number. > > Does anyone have strong preferences? > > Peter > From fernan at iib.unsam.edu.ar Thu Mar 8 10:20:31 2007 From: fernan at iib.unsam.edu.ar ('Fernan Aguero') Date: Thu, 8 Mar 2007 12:20:31 -0300 Subject: [EMBOSS] EMBOSS 4.1.0 released In-Reply-To: <45F0212A.4070204@ebi.ac.uk> References: <009601c76188$d4b7f2a0$2d01a8c0@PICO> <45F0212A.4070204@ebi.ac.uk> Message-ID: <20070308152031.GC32117@iib.unsam.edu.ar> +----[ Peter Rice (08.Mar.2007 11:52): | | Ryan Golhar wrote: | >I agree. I was also expecting the version number on the tarballs to change | >as well. At the moment, there is no way to tell they were updated... | | The embassy changes are all minor. We like to use the version numbers of | the original code so it is a little difficult to merge in the EMBOSS | version ... without making up a very long version number. | | Does anyone have strong preferences? | | Peter | +----] My original concern was about distfiles being renamed without renaming the contained directories. That is a separate problem from the other, long standing problem of EMBASSY apps distifiles being re-rolled (with changes in checksums) without version changes. 1. The problem that appeared with the 4.1.0 release: MEMENEW-0.1.0.tar.gz was renamed to MEME-0.1.0.tar.gz but extracting MEME-0.1.0.tar.gz creates a new directory named MEMENEW-0.1.0 (?) The same happens with PHYLIP/PHYLIPNEW. So, what is the correct name of this EMBASSY app? I'd like to suggest that either i) the distfiles are renamed back to MEMENEW and PHYLIPNEW or ii) the directory contained within the tarball is renamed to MEME AND PHYLIP, respectively. That is to say, keep the same name everywhere. That's what I'm suggesting for this first point. 2. The long standing problem with EMBASSY app versions: I agree with Peter in that it is good to have a version number that is in sync with the upstream sources. If the EMBASSY HMMER comes from HMMER-2.3.2 I'd like to see this reflected in the filename. But I'd also like to see reflected somewhere that this particular set of embassy distfiles are in sync with a particular version of EMBOSS. Perhaps reorganizing the layout of the FTP directories so as to have something like the following? pub/EMBOSS/ EMBOSS-3.0.0/ EMBOSS-3.0.0.tar.gz EMBASSY/ DOMAINATRIX-0.1.0.tar.gz ... EMBOSS-4.0.0/ EMBOSS-4.0.0.tar.gz EMBASSY/ DOMAINATRIX-0.1.0.tar.gz ... EMBOSS-4.1.0/ EMBOSS-4.1.0.tar.gz EMBASSY/ DOMAINATRIX-0.1.0.tar.gz ... or something like pub/EMBOSS/ EMBOSS-3.0.0/ EMBOSS-3.0.0.tar.gz EMBASSY-3.0.0/ DOMAINATRIX-0.1.0.tar.gz ... EMBOSS-4.0.0/ EMBOSS-4.0.0.tar.gz EMBASSY-4.0.0/ DOMAINATRIX-0.1.0.tar.gz ... EMBOSS-4.1.0/ EMBOSS-4.1.0.tar.gz EMBASSY-4.1.0/ DOMAINATRIX-0.1.0.tar.gz ... either way it's made clear that embassy apps are associated with a certain emboss version. Another alternative is to add some number to the original (upstream) version number that reflects this. As Peter says, adding the emboss version number (4.1.0) might make for a long version number ... but what about using the shared library (libnucleus) major version number (4 in this case)? Do the new embassy apps build and work fine against emboss-4.0.0 (libnucleus is still at v4)? Fernan From fernan at iib.unsam.edu.ar Thu Mar 8 10:27:16 2007 From: fernan at iib.unsam.edu.ar ('Fernan Aguero') Date: Thu, 8 Mar 2007 12:27:16 -0300 Subject: [EMBOSS] EMBOSS 4.1.0 released In-Reply-To: <20070308152031.GC32117@iib.unsam.edu.ar> References: <009601c76188$d4b7f2a0$2d01a8c0@PICO> <45F0212A.4070204@ebi.ac.uk> <20070308152031.GC32117@iib.unsam.edu.ar> Message-ID: <20070308152716.GD32117@iib.unsam.edu.ar> +----[ Cc emboss at emboss.open-bio.org (08.Mar.2007 12:20): | | Perhaps reorganizing the layout of the FTP directories so as | to have something like the following? | | pub/EMBOSS/ | EMBOSS-3.0.0/ | EMBOSS-3.0.0.tar.gz | EMBASSY/ | DOMAINATRIX-0.1.0.tar.gz | ... | EMBOSS-4.0.0/ | EMBOSS-4.0.0.tar.gz | EMBASSY/ | DOMAINATRIX-0.1.0.tar.gz | ... | EMBOSS-4.1.0/ | EMBOSS-4.1.0.tar.gz | EMBASSY/ | DOMAINATRIX-0.1.0.tar.gz | ... | | or something like | | pub/EMBOSS/ | EMBOSS-3.0.0/ | EMBOSS-3.0.0.tar.gz | EMBASSY-3.0.0/ | DOMAINATRIX-0.1.0.tar.gz | ... | EMBOSS-4.0.0/ | EMBOSS-4.0.0.tar.gz | EMBASSY-4.0.0/ | DOMAINATRIX-0.1.0.tar.gz | ... | EMBOSS-4.1.0/ | EMBOSS-4.1.0.tar.gz | EMBASSY-4.1.0/ | DOMAINATRIX-0.1.0.tar.gz | ... | | either way it's made clear that embassy apps are associated | with a certain emboss version. | +----] Just let me add that a similar organization seems to be in place for 'old' releases: both emboss-4.0.0, its associated embassy apps and patch files are all available under pub/EMBOSS/old/4.0.0/ In this case it's clear that these apps should be used against this emboss release. Fernan From mathog at caltech.edu Thu Mar 8 12:37:34 2007 From: mathog at caltech.edu (David Mathog) Date: Thu, 08 Mar 2007 09:37:34 -0800 Subject: [EMBOSS] EMBOSS 4.1.0 released Message-ID: > Ryan Golhar wrote: > > I agree. I was also expecting the version number on the tarballs to change > > as well. At the moment, there is no way to tell they were updated... > > The embassy changes are all minor. We like to use the version numbers of the > original code so it is a little difficult to merge in the EMBOSS version ... > without making up a very long version number. > > Does anyone have strong preferences? Use the long version number. That way it is clear at a glance what goes with what. That is, change this: EMNU-1.05.tar.gz to EMNU-1.05_for_4.1.0.tar.gz Inside the .tar it can still unpack into EMNU-1.05. The trick is making sure the right variant of EMNU (etc.) is always unpacked into the matching EMBOSS directory tree. It would probably also help if both EMBOSS and the EMBASSY programs had a top level "EMBOSS.VERSION" file, so that the Makefile for EMBASSY programs could check first thing to verify that they were in the right place, ie diff EMBOSS.VERSION .. At the moment the "4.x.y" info is relatively inaccessible in configure, Makefile, and so forth. Regards, David Mathog mathog at caltech.edu Manager, Sequence Analysis Facility, Biology Division, Caltech From beavisman44 at gmail.com Mon Mar 19 23:45:28 2007 From: beavisman44 at gmail.com (sui yun foong) Date: Tue, 20 Mar 2007 11:45:28 +0800 Subject: [EMBOSS] jemboss installation on irix 6.5.26 Message-ID: <3ea8fa90703192045m2b2b35f8r78d5aa37ac0bb646@mail.gmail.com> Greeting all, I'm currently trying to setup a jemboss client-server in sgi machine (irix 6.5.26). I have read documentation and download tomcat 5.5, emboss 4.1.0and xml-axis 1.0 (for soap). But I can't seems to set the program to run together. I am a bit new to irix system but I'm sure i can tackle if pointed the right direction. Does anybody have a detail explaination of how to install in sgi? Thanks in advance. From, foong sui yun From nyachy at yahoo.co.jp Sat Mar 24 21:03:27 2007 From: nyachy at yahoo.co.jp (=?ISO-2022-JP?B?GyRCPi5MbkRNGyhCIBskQkNSSH4bKEI=?=) Date: Sun, 25 Mar 2007 10:03:27 +0900 (JST) Subject: [EMBOSS] emma via perl/cgi system command Message-ID: <20070325010327.82506.qmail@web3704.mail.tnz.yahoo.co.jp> Dear friends I have an trouble in executing emma command. The situation is like following. I'd like to perform multiple alignment using a protein sequence fasta format as input via Perl/CGI web system. I thought the best way to achieve this objective is to use system command like ##### my $emma_com = "/usr/local/bin/emma /tmp/fasta /tmp/align /tmp/dnd"; system("$emma_com"); ##### "/tmp/fasta" is the input fasta file. "/tmp/align" and "/tmp/dnd" is supposed to be output files. This $emma_com command works when I put the command from xterm console, which means that I can get the output files.But when I use the command from Perl/CGI web system, only I can have is empty output files. At first I thought that apache can't find clustalw which is installed in /usr/local/bin but ###### system("which clustalw") ###### returns "/usr/local/bin/clustalw" This means apache detects clustalw I guess. So. my question is (1)How to get the result output file in the Perl/CGI apache web system? (2)Is there any more smart way to get the clustalw result file from command lile in Perl/CGI system? Thank you in advance. From pmr at ebi.ac.uk Sun Mar 25 08:37:42 2007 From: pmr at ebi.ac.uk (pmr at ebi.ac.uk) Date: Sun, 25 Mar 2007 13:37:42 +0100 (BST) Subject: [EMBOSS] emma via perl/cgi system command In-Reply-To: <20070325010327.82506.qmail@web3704.mail.tnz.yahoo.co.jp> References: <20070325010327.82506.qmail@web3704.mail.tnz.yahoo.co.jp> Message-ID: <51876.217.44.133.206.1174826262.squirrel@webmail.ebi.ac.uk> > Dear friends > > I have an trouble in executing emma command. > The situation is like following. > > I'd like to perform multiple alignment using a protein > sequence fasta format as input via Perl/CGI web system. > > I thought the best way to achieve this objective is to use > system command like > > ##### > my $emma_com = "/usr/local/bin/emma /tmp/fasta /tmp/align > /tmp/dnd"; > > system("$emma_com"); > ##### > > "/tmp/fasta" is the input fasta file. "/tmp/align" and > "/tmp/dnd" is supposed to be output files. > > This $emma_com command works when I put the command from > xterm console, which means that I can get the output > files.But when I use the command from Perl/CGI web system, > only I can have is empty output files. You are running emma through perl, and then through the default shell. It is possible that clustalw is not in the path when EMBOSS tries to run clustalw. (perhaps because perl/CGI web pages define the path differently, or run as a different user). To test this, add -debug to your emmma command line and look for lines in the emma.dbg file with the string "clustalw" or "CLUSTALW" You can define the full path for clustalw in the environment variable EMBOSS_CLUSTALW or as a variable in your emboss.default or ~/.embossrc file. Running with -debug will show the difference in the emma.dbg file. Hope this helps, Peter Rice From Nick at alekta.ru Fri Mar 30 02:49:57 2007 From: Nick at alekta.ru (=?koi8-r?B?98/Sz8LDz9cg7snLz8zByiD3wcTJzc/Xyd4=?=) Date: Fri, 30 Mar 2007 13:49:57 +0700 Subject: [EMBOSS] question (Error: Failed to find host 'srs.ebi.ac.uk' for database 'emblebi') Message-ID: <39C84EE794985249993243D376DEBFA5C52827@MAIL2.alekta.ru> Hello, dear Sirs! The other day I have install the EMBOSS package (version for windows). Environmen parameters are setted as required: SET EMBOSS_ROOT=D:\Emboss-MS SET EMBOSS_ACDROOT=D:\Emboss-MS\acd SET EMBOSS_DATA=D:\Emboss-MS\data Please, help me to solve a problem. Why no one's of utilities (seqret, dottup and others) do not desire work with known databases. showdb return: # Name???????? Type? ID? Qry All Comment # ============ ==== ==? === === ======= pir??????????? P??? OK? -?? -?? PIR from the EBI sw???????????? P??? OK? -?? -?? SWISSPROT from the EBI swissprot????? P??? OK? -?? -?? SWISSPROT from the EBI uni??????????? P??? OK? -?? -?? UNIPROT from the EBI uniprot??????? P??? OK? -?? -?? UNIPROT from the EBI em???????????? N??? OK? -?? -?? EMBL from the EBI emblebi??????? N??? OK? -?? -?? EMBL from the EBI gb???????????? N??? OK? -?? -?? GenBank from Infobiogen genbank??????? N??? OK? -?? -?? GenBank from Infobiogen refseq???????? N??? OK? -?? -?? REFSEQ from EBI but if I try exec: seqret emblebi:xlrhodop than get messages: Reads and writes (returns) sequences Error: Failed to find host 'srs.ebi.ac.uk' for database 'emblebi' Error: Unable to read sequence 'emblebi:xlrhodop' Died: seqret terminated: Bad value for '-sequence' and no prompt Any other db-name (from showed by showdb) work also. Please say what a problem is? In being thanksgiving for any help! From pmr at ebi.ac.uk Fri Mar 30 03:59:23 2007 From: pmr at ebi.ac.uk (Peter Rice) Date: Fri, 30 Mar 2007 08:59:23 +0100 Subject: [EMBOSS] question (Error: Failed to find host 'srs.ebi.ac.uk' for database 'emblebi') In-Reply-To: <39C84EE794985249993243D376DEBFA5C52827@MAIL2.alekta.ru> References: <39C84EE794985249993243D376DEBFA5C52827@MAIL2.alekta.ru> Message-ID: <460CC35B.30808@ebi.ac.uk> Dear Nikolai, ???????? ??????? ????????? wrote: > The other day I have install the EMBOSS package (version for windows). > > Environmen parameters are setted as required: > SET EMBOSS_ROOT=D:\Emboss-MS > SET EMBOSS_ACDROOT=D:\Emboss-MS\acd > SET EMBOSS_DATA=D:\Emboss-MS\data > > seqret emblebi:xlrhodop > Reads and writes (returns) sequences > Error: Failed to find host 'srs.ebi.ac.uk' for database 'emblebi' > Error: Unable to read sequence 'emblebi:xlrhodop' > > Please say what a problem is? The databases defined by default connect to servers here at EBI. These databases need an internet connection. If you can connect your browser to http://srs.ebi.ac.uk then EMBOSS will be able to read databases. There are some more settings you can add if, for example, you need to define an HTTP proxy. You can also install the database flat files locally and index them with dbxflat (or dbiflat). You can use any EMBOSS program with local sequence data files, or put the sequence on the command line with the syntax: seqret asis::aaaaccccggggtttt regards, Peter Rice