From xiaozhendong at gmail.com Mon May 1 12:43:14 2006 From: xiaozhendong at gmail.com (zhendong shaw) Date: Mon, 1 May 2006 09:43:14 -0700 Subject: [EMBOSS] useful hits about install EMBOSS-3.0.0 under cygwin Message-ID: Dear guys, When installed the EMBOSS-3.0.0 under cygwin, I encountered some problems. By using google searching the answers, all the problems solved. I found some guys in the mailing list asked some questions the same as I encoutered. So I list the hits about install EMBOSS-3.0.0 under cygwin, maybe they will be useful for some people. 1. Read the HowTo by Olivier Martin http://emboss.sourceforge.net/download/cygwin.html 2. If you find some compile error about "d_ino"(maybe the most popular problem ^_^), you could try as follow: *mv ajax/ajfile.c ajax/ajfile.bak mv nucleus/embdbi.c nucleus/embdbi.bak perl -pe 's/d_ino/__invalid_d_ino/g' ajax/ajfile.bak > ajax/ajfile.c perl -pe 's/d_ino/__invalid_d_ino/g' nucleus/embdbi.bak > nucleus/embdbi.c *3. Also you may find the "automake version" problem, try: * aclocal - I m4 autoconf automake -a and then make again. * Best! Zhendong Shaw From ajb at ebi.ac.uk Tue May 2 04:01:52 2006 From: ajb at ebi.ac.uk (ajb at ebi.ac.uk) Date: Tue, 2 May 2006 09:01:52 +0100 (BST) Subject: [EMBOSS] useful hits about install EMBOSS-3.0.0 under cygwin In-Reply-To: References: Message-ID: <34986.81.98.244.247.1146556912.squirrel@webmail.ebi.ac.uk> Thanks for the post. Some quick comments though. > 1. Read the HowTo by Olivier Martin > http://emboss.sourceforge.net/download/cygwin.html Although there is nothing really inaccurate about the HowTo it does imply that --without-x is essential. X windows works fine as long as you've installed it under cygwin. It also suggests using CFLAGS=-s to the configure, presumably to strip the binaries. This should not be required as the latest cygwin distributions automatically strip the binaries. > 2. If you find some compile error about "d_ino"(maybe the most popular > problem ^_^), you could try as follow: > *mv ajax/ajfile.c ajax/ajfile.bak > mv nucleus/embdbi.c nucleus/embdbi.bak > perl -pe 's/d_ino/__invalid_d_ino/g' ajax/ajfile.bak > ajax/ajfile.c > perl -pe 's/d_ino/__invalid_d_ino/g' nucleus/embdbi.bak > > nucleus/embdbi.c This would indeed work but you might as well apply all the fixes from: ftp://emboss.open-bio.org/pub/EMBOSS/fixes/ which fix this and other issues. > *3. Also you may find the "automake version" problem, try: > * aclocal - I m4 > autoconf > automake -a > and then make again. > * I don't think anyone has ever reported "the automake version" problem so I'm unsure what is meant here. HTH Alan From stefan.rensing at biologie.uni-freiburg.de Thu May 4 08:35:22 2006 From: stefan.rensing at biologie.uni-freiburg.de (Stefan Rensing) Date: Thu, 04 May 2006 14:35:22 +0200 Subject: [EMBOSS] sixpack Message-ID: <4459F50A.1080503@biologie.uni-freiburg.de> Dear all, is there a way to create a sixpack output that contains the n.a. ORFs and/or their coordinates in the fasta description line? Also, is there an option to output the longest ORF only? And yes, both would be useful. ;) Cheers, Stefan -- Dr. Stefan Rensing, Group Leader Computational Biology Plant Biotechnology, Faculty of Biology, University of Freiburg Schaenzlestr. 1, D-79104 Freiburg, Fon: +49 761 203-6974, Fax: -6945 http://www.plant-biotech.net/ http://www.cosmoss.org/ stefan.rensing at biologie.uni-freiburg.de "I want you to take him from his hotel, drug him, put him in the front of a car, and run a truck into it at 50 mph." From jison at ebi.ac.uk Thu May 4 10:49:56 2006 From: jison at ebi.ac.uk (Jon Ison) Date: Thu, 4 May 2006 15:49:56 +0100 (BST) Subject: [EMBOSS] sixpack In-Reply-To: <4459F50A.1080503@biologie.uni-freiburg.de> References: <4459F50A.1080503@biologie.uni-freiburg.de> Message-ID: <38707.172.31.70.94.1146754196.squirrel@webmail.ebi.ac.uk> Hi Stefan The short answer is no, but both would be easy to easy to implement as ACD options - perhaps for the next release. Please send an example of the fasta description line you have in mind. Cheers Jon > Dear all, > > is there a way to create a sixpack output that contains the n.a. ORFs > and/or their coordinates in the fasta description line? > > Also, is there an option to output the longest ORF only? > > And yes, both would be useful. ;) > > Cheers, Stefan > > > -- > Dr. Stefan Rensing, Group Leader Computational Biology > Plant Biotechnology, Faculty of Biology, University of Freiburg > Schaenzlestr. 1, D-79104 Freiburg, Fon: +49 761 203-6974, Fax: -6945 > http://www.plant-biotech.net/ http://www.cosmoss.org/ > stefan.rensing at biologie.uni-freiburg.de > > "I want you to take him from his hotel, drug him, put him in the > front of a car, and run a truck into it at 50 mph." > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From d.gatherer at vir.gla.ac.uk Fri May 5 04:41:40 2006 From: d.gatherer at vir.gla.ac.uk (Derek Gatherer) Date: Fri, 05 May 2006 09:41:40 +0100 Subject: [EMBOSS] sixpack Message-ID: <6.2.3.4.1.20060505094057.02ad0ba0@lenzie.gla.ac.uk> Hi Jon et al First of all congrats on the new EMBOSS funding. Stefan has pointed to an important problem - we really need a good _flexible_ translator. The current bunch are powerful and fast but too rigid and in some cases don't appear to do what they say on the tin. See my previous posts: http://emboss.open-bio.org/pipermail/emboss/2006-February/002433.html http://emboss.open-bio.org/pipermail/emboss/2006-February/002434.html I have also some private correspondence with Alan, and Gary Williams, about these issues which I'd be happy to forward with their permission. What we need is an app that enables you to say: take this sequence or sequences, translate them from various x to various y, specifying strand, and give me output format z. Currently we don't quite have that. If you could tweak the existing showseq/prettyseq/sixpack, it would be great. Cheers Derek At 15:49 04/05/2006, you wrote: >Hi Stefan > >The short answer is no, but both would be easy to easy to >implement as ACD options - perhaps for the next release. >Please send an example of the fasta description line you >have in mind. > >Cheers > >Jon > > > > > Dear all, > > > > is there a way to create a sixpack output that contains the n.a. ORFs > > and/or their coordinates in the fasta description line? > > > > Also, is there an option to output the longest ORF only? > > > > And yes, both would be useful. ;) > > > > Cheers, Stefan > > > > > > -- > > Dr. Stefan Rensing, Group Leader Computational Biology > > Plant Biotechnology, Faculty of Biology, University of Freiburg > > Schaenzlestr. 1, D-79104 Freiburg, Fon: +49 761 203-6974, Fax: -6945 > > http://www.plant-biotech.net/ http://www.cosmoss.org/ > > stefan.rensing at biologie.uni-freiburg.de > > > > "I want you to take him from his hotel, drug him, put him in the > > front of a car, and run a truck into it at 50 mph." > > _______________________________________________ > > EMBOSS mailing list > > EMBOSS at lists.open-bio.org > > http://lists.open-bio.org/mailman/listinfo/emboss > > > > >_______________________________________________ >EMBOSS mailing list >EMBOSS at lists.open-bio.org >http://lists.open-bio.org/mailman/listinfo/emboss From Emily.Wong at uts.edu.au Tue May 9 02:23:19 2006 From: Emily.Wong at uts.edu.au (Emily Wong) Date: Tue, 09 May 2006 16:23:19 +1000 Subject: [EMBOSS] emboss on intel macs Message-ID: <44603557.1070302@uts.edu.au> Hi, I'm trying to get emboss running on my intel macbook pro. Will there be a release soon that will allow this? Are there any known workarounds to this problem? Thanks in advance -- Emily Wong Application Support Specialist (Bioinformatics) Computational Research Support Unit University of Technology, Sydney, Australia PO Box 123 Broadway NSW 2007 phone: +61 9514 1723 http://crsu.science.uts.edu.au From ajb at ebi.ac.uk Tue May 9 03:09:58 2006 From: ajb at ebi.ac.uk (ajb at ebi.ac.uk) Date: Tue, 9 May 2006 08:09:58 +0100 (BST) Subject: [EMBOSS] emboss on intel macs In-Reply-To: <44603557.1070302@uts.edu.au> References: <44603557.1070302@uts.edu.au> Message-ID: <33314.172.31.100.160.1147158598.squirrel@webmail.ebi.ac.uk> Hello, As long as you apply all the fixes from: ftp://emboss.open-bio.org/pub/EMBOSS/fixes/ then you should be OK. Check the README there for details. HTH Alan > Hi, > > I'm trying to get emboss running on my intel macbook pro. Will there be > a release soon that will allow this? Are there any known workarounds to > this problem? > > Thanks in advance > > -- > Emily Wong > Application Support Specialist (Bioinformatics) > Computational Research Support Unit > University of Technology, Sydney, Australia > PO Box 123 Broadway NSW 2007 > phone: +61 9514 1723 > http://crsu.science.uts.edu.au > > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From kvddrift at earthlink.net Tue May 9 06:22:29 2006 From: kvddrift at earthlink.net (Koen van der Drift) Date: Tue, 9 May 2006 06:22:29 -0400 Subject: [EMBOSS] emboss on intel macs In-Reply-To: <44603557.1070302@uts.edu.au> References: <44603557.1070302@uts.edu.au> Message-ID: <38B3B778-EF01-4FE4-9D1A-AFD6A76E3B7B@earthlink.net> On May 9, 2006, at 2:23 AM, Emily Wong wrote: > Hi, > > I'm trying to get emboss running on my intel macbook pro. Will > there be > a release soon that will allow this? Are there any known > workarounds to > this problem? You can also use fink to install EMBOSS on your mac. As far as I know there are no issues with intel processors for version 3.0.0. - Koen. From Emily.Wong at uts.edu.au Wed May 10 23:09:29 2006 From: Emily.Wong at uts.edu.au (Emily Wong) Date: Thu, 11 May 2006 13:09:29 +1000 Subject: [EMBOSS] oddcomp result wrong Message-ID: <4462AAE9.40105@uts.edu.au> Hi, Thanks for the information about installing emboss on intel macs earlier. Unfortunately I've run into another problem... My aim is to modify oddcomp to provide some additional functionalities. I have successfully installed emboss from fink. Also I have successfully compiled the source code of oddcomp (obtained from emboss-3.0.0 at emboss). However I get different results to the same input data when I run oddcomp from the fink install compared with running oddcomp from the lone compilation. The results for the lone compile is wrong. My compseq-like input is here: Word size 1 Total count 0 A 5 _Output of fink installed oddcomp_ # # Output from 'oddcomp' # # The Expected frequencies are taken from the file: input_like_COMPSEQ.txt # # Word size: 1 *longsequenceAs* # END # _Output of self compiled oddcomp_ # # Output from 'oddcomp' # # The Expected frequencies are taken from the file: input_like_COMPSEQ.txt # # Word size: 1 * longseqnoAs* # END # Any sugestions as to the cause of the problem? Thanks in advance, Emily From tmargus at ebc.ee Thu May 11 08:21:28 2006 From: tmargus at ebc.ee (=?iso-8859-1?Q?T=F5nu_Margus?=) Date: Thu, 11 May 2006 15:21:28 +0300 Subject: [EMBOSS] Questions about emboss bugfixses, rpm installations and dbxflat In-Reply-To: <33314.172.31.100.160.1147158598.squirrel@webmail.ebi.ac.uk> Message-ID: <000601c674f5$7fdae7e0$901728c1@MULTIVITAMIIN> Hi, How to repair/update rpm installation? I fetched and installed version 3.0.0.2 from http://serine.umdnj.edu/~golharam/biorpms/ Now there is version 3.0.0.3 I don't know are there dbxflat bugs are fixed or not? When I tried to index 16S rRNA sequences (genbank format v9.38 ) fetched from RDP I got an error after ~2h indexing: Database b+tree indexing for flat file databases Processing file ./release9_38_unaligned.gb EMBOSS An error in ajindex.c at line 526: Seek error in ajBtreeCachefetch Command line was dbxflat -dbresource GB -idformat Genbank -directory . -filenames *.gb -dbname SSU9 -release 9.38 -date 11.05.06 -fields id,acc,sv,des,key,org Any help is welcome T?nu Margus -----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: Tuesday, May 09, 2006 10:10 AM To: Emily Wong Cc: emboss at lists.open-bio.org Subject: Re: [EMBOSS] emboss on intel macs Hello, As long as you apply all the fixes from: ftp://emboss.open-bio.org/pub/EMBOSS/fixes/ then you should be OK. Check the README there for details. HTH Alan > Hi, > > I'm trying to get emboss running on my intel macbook pro. Will there be > a release soon that will allow this? Are there any known workarounds to > this problem? > > Thanks in advance > > -- > Emily Wong > Application Support Specialist (Bioinformatics) > Computational Research Support Unit > University of Technology, Sydney, Australia > PO Box 123 Broadway NSW 2007 > phone: +61 9514 1723 > http://crsu.science.uts.edu.au > > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > _______________________________________________ EMBOSS mailing list EMBOSS at lists.open-bio.org http://lists.open-bio.org/mailman/listinfo/emboss From aheyland at ufl.edu Sun May 14 23:15:07 2006 From: aheyland at ufl.edu (Andreas Heyland) Date: Sun, 14 May 2006 20:15:07 -0700 Subject: [EMBOSS] problems installing EMBOSS under Cygwin Message-ID: <4467F23B.7010403@ufl.edu> Hi, I download Emboss package and tried to install it under cygwin. I keep getting the error message cc1.exe - Entry Point not found The procedure entry point_impure_ptr could not be located in the dynamic link library cygwin1.dll when I try to run the: "./configure --without-x CFLAGS=-s" command. I am using: cygwin: 1.5.19-4 gcc: 3.4.4-1 It seems that my cygwin installation might not have worked, i.e. there is an incompatibility issue between GCC and Cygwin. Any suggestions? Thanks a lot Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: aheyland.vcf Type: text/x-vcard Size: 437 bytes Desc: not available Url : http://lists.open-bio.org/pipermail/emboss/attachments/20060514/19ab6e48/attachment.vcf From stefan.rensing at biologie.uni-freiburg.de Wed May 17 08:16:35 2006 From: stefan.rensing at biologie.uni-freiburg.de (Stefan Rensing) Date: Wed, 17 May 2006 14:16:35 +0200 Subject: [EMBOSS] strange frames Message-ID: <446B1423.3010202@biologie.uni-freiburg.de> Dear all, apparently, sixpack/transeq handle frames on the reverse complementary strand different than e.g. BLAST does (and different than I would have expected). Using sixpack/transeq to translate this sequence: AGATGACTCCGTTGCAGTTGGCTACTGCTTATATTGCATCAGGGGGTGTGCTTCA Sixpack output: R * L R C S W L L L I L H Q G V C F X F1 D D S V A V G Y C L Y C I R G C A S F2 M T P L Q L A T A Y I A S G G V L F3 1 AGATGACTCCGTTGCAGTTGGCTACTGCTTATATTGCATCAGGGGGTGTGCTTCA 55 ----:----|----:----|----:----|----:----|----:----|----:----| 1 TCTACTGAGGCAACGTCAACCGATGACGAATATAACGTAGTCCCCCACACGAAGT 55 L H S R Q L Q S S S I N C * P T H K F6 X I V G N C N A V A * I A D P P T S * F5 S S E T A T P * Q K Y Q M L P H A E F4 Transeq output: Frame -2: >>test_4 EAHPLMQYKQ*PTATESS Frame -1: >>test_5 *STPPDAI*AVANCNGVIX Frame -3: >>test_6 KHTP*CNISSSQLQRSHL IMHO, it should be: Frame 1 = +1 Frame 2 = +2 Frame 3 = +3 Frame 4 = -1 Frame 5 = -2 Frame 6 = -3 However, in this case, sixpack and transeq return: Frame 1 = +1 Frame 2 = +2 Frame 3 = +3 Frame 4 = -2 Frame 5 = -1 Frame 6 = -3 Is this the expected behaviour, and if so, why? Regards, Stefan -- Dr. Stefan Rensing, Group Leader Computational Biology Plant Biotechnology, Faculty of Biology, University of Freiburg Schaenzlestr. 1, D-79104 Freiburg, Fon: +49 761 203-6974, Fax: -6945 http://www.plant-biotech.net/ http://www.cosmoss.org/ stefan.rensing at biologie.uni-freiburg.de "Mr. Musgrave, please don't interrupt me when I'm asking rhetorical questions." From jison at ebi.ac.uk Wed May 17 09:17:38 2006 From: jison at ebi.ac.uk (Jon Ison) Date: Wed, 17 May 2006 14:17:38 +0100 (BST) Subject: [EMBOSS] problems installing EMBOSS under Cygwin In-Reply-To: <4467F23B.7010403@ufl.edu> References: <4467F23B.7010403@ufl.edu> Message-ID: <46444.172.31.100.168.1147871858.squirrel@webmail.ebi.ac.uk> Hi Andreas I don't think you got a reply to this. I've had a trawl through the messages to the EMBOSS lists ... perhaps some of the following will help (I'm no expert on cygwin) ? A port of EMBOSS 3 to windows will be made available soonish - so you'll never need to use cygwin again. Cheers Jon 1. Apply all the fixes from: ftp://emboss.open-bio.org/pub/EMBOSS/fixes/ 2. Update cygwin Problems can arise from using older versions of cygwin, or selective ("pick and choose") installations of cygwin. Download a new, completely fresh copy of cygwin (the cygwin updating mechanism has proved unsatisfactory) and accept the default packages; best to get a pretty complete set of cygwin packages. If you decide to install a fresh copy then be sure to delete any cygwin registry entries after deleting the cygwin directory: see (1) DELETING CYGWIN below (email emboss-bug if the registry fixing script cygtidy.reg is not available on the website). Your cygwin installation directory should be around 1.5 - 2Gb in size. 3. Read the instructions on the web, but see the corrections (1) to that text, which haven't made it onto the website yet: http://emboss.sourceforge.net/download/cygwin.html The instructions imply that --without-x is essential, whereas X windows works fine as long as you've installed it under cygwin. It also suggests using CFLAGS=-s to the configure, presumably to strip the binaries. This should not be required as the latest cygwin distributions automatically strip the binaries. (1) CORRECTIONS / ADDITIONS 1) install cygwin (http://www.cygwin.com). That's surely possible to compile without it, but it's making everything easier. You'll have to install the base part plus the devel/make and devel/gcc packages. Before installing, see the first line of 2). Then you'll have a linux-like command-line with tools like gcc and make. The paths are then made according to the place where you installed cygwin (ex: c\cygwin becomes /). Notes from the EMBOSS developers: We recommend doing a full cygwin installation or problems may arise with creating the DLLs. 3) quit and restart the cygwin bash (you'll start in your "FirstName_LastName" directory, you can delete "FirstName LastName"), download the EMBOSS package and cd to the place where you uncompressed it. Follow the usual procedure: ./configure --without-x CFLAGS=-s make make install --without-x because at one point the X11 library is required (even if useless) and without this flag the compilation would stop. CFLAGS=-s to make the binaries smaller. Notes from the EMBOSS developers: If you have done a reasonably full cygwin installation then you can leave out the --without-x switch. X will work perfectly well if installed. Also, with later cygwin installations, you can omit the CFLAGS=-s switch. We also recommend providing a prefix switch to install EMBOSS in a specified location. The configuration then becomes: ./configure --prefix=/usr/local/emboss make make install After installation make sure that /usr/local/emboss/bin is in your PATH. DELETING CYGWIN To delete cygwin you should drag your cygwin installation directory to the recycle bin. Then delete any cygwin entries from the 'start' menu. You should then download the file cygtidy.reg and double-click it to remove any cygwin registry entries. > Hi, I download Emboss package and tried to install it under cygwin. I > keep getting the error message cc1.exe - Entry Point not found The > procedure entry point_impure_ptr could not be located in the dynamic > link library cygwin1.dll when I try to run the: "./configure --without-x > CFLAGS=-s" command. I am using: > cygwin: 1.5.19-4 > gcc: 3.4.4-1 > It seems that my cygwin installation might not have worked, i.e. there > is an incompatibility issue between GCC and Cygwin. Any suggestions? > Thanks a lot > Andreas > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From sbassi at gmail.com Wed May 17 11:04:15 2006 From: sbassi at gmail.com (Sebastian Bassi) Date: Wed, 17 May 2006 12:04:15 -0300 Subject: [EMBOSS] problems installing EMBOSS under Cygwin In-Reply-To: <46444.172.31.100.168.1147871858.squirrel@webmail.ebi.ac.uk> References: <4467F23B.7010403@ufl.edu> <46444.172.31.100.168.1147871858.squirrel@webmail.ebi.ac.uk> Message-ID: On 5/17/06, Jon Ison wrote: > I don't think you got a reply to this. I've had a trawl through the > messages to the EMBOSS lists ... perhaps some of the following will > help (I'm no expert on cygwin) ? A port of EMBOSS 3 to windows will > be made available soonish - so you'll never need to use cygwin again. There is another solution: Use EMBOSS w/o install Linux, with a LiveCD. That is why I included EMBOSS 3.0 in DNALinux Desktop edition: http://www.dnalinux.com It is free as a GPL :) -- Bioinformatics news: http://www.bioinformatica.info Lriser: http://www.linspire.com/lraiser_success.php?serial=318 From Staffan.X.Bergh at astrazeneca.com Thu May 18 06:48:31 2006 From: Staffan.X.Bergh at astrazeneca.com (Staffan.X.Bergh at astrazeneca.com) Date: Thu, 18 May 2006 12:48:31 +0200 Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Message-ID: Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box with RedHat Enterprise Server 3 (not the latest, I know, but corporate standards, etc, etc, ...) The error message from make/ld is: > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when searching for -lX11 > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when searching for -lX11 > /usr/bin/ld: cannot find -lX11 The X version that RHES3 ships with is 4.3.0 > $ X -version > XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: 4.3.0-68.EL) > Release Date: 15 August 2003 > (and a lot more, deleted by the RSSPBW [Royal Swedish Society for the Preservation of BandWidth]). I couldn't find answer to this in the docs, nor in the mail archives, and I guess that I'll end up having to upgrade X, but it would be nice if someone could confirm that that would solve the problem before I try? Or is this something else? cheers /staffan _____________________________________________________________ Staffan Bergh, PhD Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D B209:537, SE-151 81 S?dert?lje, Sweden Tel: +46 8 553 215 72 staffan.x.bergh at astrazeneca.com From Marc.Logghe at DEVGEN.com Thu May 18 08:30:00 2006 From: Marc.Logghe at DEVGEN.com (Marc Logghe) Date: Thu, 18 May 2006 14:30:00 +0200 Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Message-ID: <0C528E3670D8CE4B8E013F6749231AA6746DD0@ANTARESIA.be.devgen.com> I ran into the same issue a while ago. If I recall well you had to run configure with the option '--without-x'. HTH, Marc > -----Original Message----- > From: emboss-bounces at lists.open-bio.org > [mailto:emboss-bounces at lists.open-bio.org] On Behalf Of > Staffan.X.Bergh at astrazeneca.com > Sent: Thursday, May 18, 2006 12:49 PM > To: emboss at emboss.open-bio.org > Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? > > Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box > with RedHat Enterprise Server 3 (not the latest, I know, but > corporate standards, etc, etc, ...) > > The error message from make/ld is: > > > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when > > searching > for -lX11 > > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when > > searching > for -lX11 > > /usr/bin/ld: cannot find -lX11 > > The X version that RHES3 ships with is 4.3.0 > > > $ X -version > > XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: > > 4.3.0-68.EL) Release Date: 15 August 2003 (and a lot more, > deleted by > > the RSSPBW [Royal Swedish Society for the > Preservation of BandWidth]). > > I couldn't find answer to this in the docs, nor in the mail > archives, and I guess that I'll end up having to upgrade X, > but it would be nice if someone could confirm that that would > solve the problem before I try? > Or is this something else? > > cheers /staffan > _____________________________________________________________ > Staffan Bergh, PhD > Informatics and Logistics, Molecular Pharmacology, > AstraZeneca R&D B209:537, SE-151 81 S?dert?lje, Sweden > Tel: +46 8 553 215 72 > staffan.x.bergh at astrazeneca.com > > > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From ajb at ebi.ac.uk Thu May 18 10:21:24 2006 From: ajb at ebi.ac.uk (ajb at ebi.ac.uk) Date: Thu, 18 May 2006 15:21:24 +0100 (BST) Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? In-Reply-To: References: Message-ID: <47180.81.98.244.247.1147962084.squirrel@webmail.ebi.ac.uk> If you do a: setenv CFLAGS "-D__amd64__" (4 underscores in all!) before configuring then that will stop the segfaulting on opterons. You should at least 'make clean' before configuring again. The above will not be required in the next release of EMBOSS. Alan > Hi again, > > the machine is an HP DL385 med 2xOpteron 2.8GHz CPUs ... > and yes, I've checked for the 64bit libraries, too. > > To add to my woes, configuring "--without-x" (thanks Mark > for the pointer!) and compiling (w/o warnings but for two > incompatible pointers in ajfmt.c) gives binaries that segfault > immediately. The config.log has some more > > '/usr/bin/ld: skipping incompatible' library, > > but now for libz, libm and libc in the conftest for inflateEnd > in -lz ... So I tried turning configuring w/o the pngdriver, but > still get segfault'ing binaries. > > So I'm at a bit of a loss about what to do right > now - except force our IS/IT dept to upgrade RedHat ... > > If anyone want to see the output from configure and/or make you're > welcome (but no sniffing around ;-) > > cheers /staffan > > -----Original Message----- > From: ajb at ebi.ac.uk [mailto:ajb at ebi.ac.uk] > Sent: 18 May 2006 15:17 > To: Staffan.X.Bergh at astrazeneca.com > Subject: RE: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? > > > Hi Staffan, > > Oh dear. Is it an AMD64 or EM64T by any chance as that error might make > sense > if it was trying to combine 32bit and 64 bit libraries. Noone has > reported it before though - but I expect few of our users have > RHES. If it is a 64bit m/c you will, of course, have checked that > both 32bit and 64bit libs/devel files have been installed. > > Also, as you're corporate, there is patently no chance of a sniff > around from this end. > > It looks like an X11 upgrade might be worth a try. > > Cheers > Alan > > > > > > > > >> Hi Alan, >> >> I do have the devel package installed ... >> >>> $ rpm -q -f /usr/X11R6/lib/libX11.a -i >>> Name : XFree86-devel Relocations: (not >>> relocatable) >>> Version : 4.3.0 Vendor: Red Hat, Inc. >>> Release : 68.EL Build Date: Sun 22 Aug 2004 >> 04:22:47 PM CEST >>> Install Date: Sun 25 Sep 2005 01:05:22 AM CEST Build Host: >> tweety.build.redhat.com >>> Group : Development/Libraries Source RPM: >> XFree86-4.3.0-68.EL.src.rpm >>> Size : 13528038 License: >> MIT/X11/XFree86/many others >>> Signature : DSA/SHA1, Mon 23 Aug 2004 07:58:10 PM CEST, Key ID >> 219180cddb42a60e >>> Packager : Red Hat, Inc. >>> URL : http://www.xfree86.org >>> Summary : X11R6 static libraries, headers and programming man >>> pages. >>> Description : >>> XFree86-devel includes the libraries, header files and documentation >>> needed for developing applications for the X window system. >> >> and it verifies OK, so it's not broken in some mysterious way ... >> >> cheers /staffan >> >> -----Original Message----- >> From: ajb at ebi.ac.uk [mailto:ajb at ebi.ac.uk] >> Sent: 18 May 2006 13:26 >> To: Staffan.X.Bergh at astrazeneca.com >> Cc: emboss-bug at emboss.open-bio.org >> Subject: Re: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? >> >> >> Hi Staffan, >> >> First check that you have the XFree86-devel RPM installed. >> If not, then install it, do a make clean and then configure >> again. >> >> Let us know whether that solves it. I've never seen that error >> in quite that form before but the XFree86-devel RPM is essential. >> >> Alan Bleasby >> EBI >> >> PS: CCing to the rest of the developers >> >> >>> Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box with >>> RedHat >>> Enterprise Server 3 (not >>> the latest, I know, but corporate standards, etc, etc, ...) >>> >>> The error message from make/ld is: >>> >>>> /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when >>>> searching >>> for -lX11 >>>> /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when >>>> searching >>> for -lX11 >>>> /usr/bin/ld: cannot find -lX11 >>> >>> The X version that RHES3 ships with is 4.3.0 >>> >>>> $ X -version >>>> XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: >>>> 4.3.0-68.EL) >>>> Release Date: 15 August 2003 >>>> (and a lot more, deleted by the RSSPBW [Royal Swedish Society for the >>> Preservation of BandWidth]). >>> >>> I couldn't find answer to this in the docs, nor in the mail archives, >>> and >>> I >>> guess that I'll end up having to >>> upgrade X, but it would be nice if someone could confirm that that >>> would >>> solve the problem before I try? >>> Or is this something else? >>> >>> cheers /staffan >>> _____________________________________________________________ >>> Staffan Bergh, PhD >>> Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D >>> B209:537, SE-151 81 S?dert?lje, Sweden >>> Tel: +46 8 553 215 72 >>> staffan.x.bergh at astrazeneca.com >>> >>> >>> _______________________________________________ >>> EMBOSS mailing list >>> EMBOSS at lists.open-bio.org >>> http://lists.open-bio.org/mailman/listinfo/emboss >>> >>> >> >> >> > > From Staffan.X.Bergh at astrazeneca.com Thu May 18 10:11:13 2006 From: Staffan.X.Bergh at astrazeneca.com (Staffan.X.Bergh at astrazeneca.com) Date: Thu, 18 May 2006 16:11:13 +0200 Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Message-ID: Hi again, the machine is an HP DL385 med 2xOpteron 2.8GHz CPUs ... and yes, I've checked for the 64bit libraries, too. To add to my woes, configuring "--without-x" (thanks Mark for the pointer!) and compiling (w/o warnings but for two incompatible pointers in ajfmt.c) gives binaries that segfault immediately. The config.log has some more '/usr/bin/ld: skipping incompatible' library, but now for libz, libm and libc in the conftest for inflateEnd in -lz ... So I tried turning configuring w/o the pngdriver, but still get segfault'ing binaries. So I'm at a bit of a loss about what to do right now - except force our IS/IT dept to upgrade RedHat ... If anyone want to see the output from configure and/or make you're welcome (but no sniffing around ;-) cheers /staffan -----Original Message----- From: ajb at ebi.ac.uk [mailto:ajb at ebi.ac.uk] Sent: 18 May 2006 15:17 To: Staffan.X.Bergh at astrazeneca.com Subject: RE: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Hi Staffan, Oh dear. Is it an AMD64 or EM64T by any chance as that error might make sense if it was trying to combine 32bit and 64 bit libraries. Noone has reported it before though - but I expect few of our users have RHES. If it is a 64bit m/c you will, of course, have checked that both 32bit and 64bit libs/devel files have been installed. Also, as you're corporate, there is patently no chance of a sniff around from this end. It looks like an X11 upgrade might be worth a try. Cheers Alan > Hi Alan, > > I do have the devel package installed ... > >> $ rpm -q -f /usr/X11R6/lib/libX11.a -i >> Name : XFree86-devel Relocations: (not >> relocatable) >> Version : 4.3.0 Vendor: Red Hat, Inc. >> Release : 68.EL Build Date: Sun 22 Aug 2004 > 04:22:47 PM CEST >> Install Date: Sun 25 Sep 2005 01:05:22 AM CEST Build Host: > tweety.build.redhat.com >> Group : Development/Libraries Source RPM: > XFree86-4.3.0-68.EL.src.rpm >> Size : 13528038 License: > MIT/X11/XFree86/many others >> Signature : DSA/SHA1, Mon 23 Aug 2004 07:58:10 PM CEST, Key ID > 219180cddb42a60e >> Packager : Red Hat, Inc. >> URL : http://www.xfree86.org >> Summary : X11R6 static libraries, headers and programming man pages. >> Description : >> XFree86-devel includes the libraries, header files and documentation >> needed for developing applications for the X window system. > > and it verifies OK, so it's not broken in some mysterious way ... > > cheers /staffan > > -----Original Message----- > From: ajb at ebi.ac.uk [mailto:ajb at ebi.ac.uk] > Sent: 18 May 2006 13:26 > To: Staffan.X.Bergh at astrazeneca.com > Cc: emboss-bug at emboss.open-bio.org > Subject: Re: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? > > > Hi Staffan, > > First check that you have the XFree86-devel RPM installed. > If not, then install it, do a make clean and then configure > again. > > Let us know whether that solves it. I've never seen that error > in quite that form before but the XFree86-devel RPM is essential. > > Alan Bleasby > EBI > > PS: CCing to the rest of the developers > > >> Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box with >> RedHat >> Enterprise Server 3 (not >> the latest, I know, but corporate standards, etc, etc, ...) >> >> The error message from make/ld is: >> >>> /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when >>> searching >> for -lX11 >>> /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when >>> searching >> for -lX11 >>> /usr/bin/ld: cannot find -lX11 >> >> The X version that RHES3 ships with is 4.3.0 >> >>> $ X -version >>> XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: 4.3.0-68.EL) >>> Release Date: 15 August 2003 >>> (and a lot more, deleted by the RSSPBW [Royal Swedish Society for the >> Preservation of BandWidth]). >> >> I couldn't find answer to this in the docs, nor in the mail archives, >> and >> I >> guess that I'll end up having to >> upgrade X, but it would be nice if someone could confirm that that would >> solve the problem before I try? >> Or is this something else? >> >> cheers /staffan >> _____________________________________________________________ >> Staffan Bergh, PhD >> Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D >> B209:537, SE-151 81 S?dert?lje, Sweden >> Tel: +46 8 553 215 72 >> staffan.x.bergh at astrazeneca.com >> >> >> _______________________________________________ >> EMBOSS mailing list >> EMBOSS at lists.open-bio.org >> http://lists.open-bio.org/mailman/listinfo/emboss >> >> > > > From golharam at umdnj.edu Thu May 18 12:00:29 2006 From: golharam at umdnj.edu (Ryan Golhar) Date: Thu, 18 May 2006 12:00:29 -0400 Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? In-Reply-To: Message-ID: <00a501c67a94$2f7ac000$2f01a8c0@GOLHARMOBILE1> That's interesting. We are running Enterprise v3 here, both workstation and server and I haven't seen this. We are also running with the latest patches from RedHat. We've also built RPMs specifically for RHEL v3 for EMBOSS (and other applications). You can get them from here: http://serine.umdnj.edu/~golharam/biorpms Ryan -----Original Message----- From: emboss-bounces at lists.open-bio.org [mailto:emboss-bounces at lists.open-bio.org] On Behalf Of Staffan.X.Bergh at astrazeneca.com Sent: Thursday, May 18, 2006 6:49 AM To: emboss at emboss.open-bio.org Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box with RedHat Enterprise Server 3 (not the latest, I know, but corporate standards, etc, etc, ...) The error message from make/ld is: > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when > searching for -lX11 > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when > searching for -lX11 > /usr/bin/ld: cannot find -lX11 The X version that RHES3 ships with is 4.3.0 > $ X -version > XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: 4.3.0-68.EL) > Release Date: 15 August 2003 > (and a lot more, deleted by the RSSPBW [Royal Swedish Society for the Preservation of BandWidth]). I couldn't find answer to this in the docs, nor in the mail archives, and I guess that I'll end up having to upgrade X, but it would be nice if someone could confirm that that would solve the problem before I try? Or is this something else? cheers /staffan _____________________________________________________________ Staffan Bergh, PhD Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D B209:537, SE-151 81 S?dert?lje, Sweden Tel: +46 8 553 215 72 staffan.x.bergh at astrazeneca.com _______________________________________________ EMBOSS mailing list EMBOSS at lists.open-bio.org http://lists.open-bio.org/mailman/listinfo/emboss From Staffan.X.Bergh at astrazeneca.com Fri May 19 09:47:10 2006 From: Staffan.X.Bergh at astrazeneca.com (Staffan.X.Bergh at astrazeneca.com) Date: Fri, 19 May 2006 15:47:10 +0200 Subject: [EMBOSS] Solved! Was: X11 version requirement for EMBOSS 3.0.0? Message-ID: OK, got the X problem worked around now. Thanks all who answered (and special thanks to Alan for not blasting my ears off for missing the FAQ about the Opterons ...) In summary: On our bright new box with 2xOpteron processors, and RedHat Enterprise Server 3.0, I was trying to compile EMBOSS-3.0.0. I got strange errors from the linker, like skipping incompatible "/usr/X11R6/lib/libX11.a" Solution was to upgrade the X libraries: version 4.6.0 Xprg.tgz file from the Xfree86.org ftp site, install in a convenient location and then point configure at the new X libraries: ./configure --x-libraries=/usr/local/X11/lib64/ \ --x-includes=/usr/local/X11/include/ That solved the compilation problem. Setting Pseudocolor in Exceed then solved next (and last, I hope) problem: Error in XCreatePixmap: BadDrawable (invalid Pixmap or Window parameter) The actual reason for the linking problem remains unclear, but I note that the rpm we had installed (that came with RHEL) had both static and shared 32bit libraries, but only shared 64bit libraries ... so maybe something not quite standard happened at RedHat when they made the rpm ... cheers /staffan _____________________________________________________________ Staffan Bergh, PhD Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D B209:537, SE-151 81 S?dert?lje, Sweden Tel: +46 8 553 215 72 staffan.x.bergh at astrazeneca.com From xiaozhendong at gmail.com Mon May 1 16:43:14 2006 From: xiaozhendong at gmail.com (zhendong shaw) Date: Mon, 1 May 2006 09:43:14 -0700 Subject: [EMBOSS] useful hits about install EMBOSS-3.0.0 under cygwin Message-ID: Dear guys, When installed the EMBOSS-3.0.0 under cygwin, I encountered some problems. By using google searching the answers, all the problems solved. I found some guys in the mailing list asked some questions the same as I encoutered. So I list the hits about install EMBOSS-3.0.0 under cygwin, maybe they will be useful for some people. 1. Read the HowTo by Olivier Martin http://emboss.sourceforge.net/download/cygwin.html 2. If you find some compile error about "d_ino"(maybe the most popular problem ^_^), you could try as follow: *mv ajax/ajfile.c ajax/ajfile.bak mv nucleus/embdbi.c nucleus/embdbi.bak perl -pe 's/d_ino/__invalid_d_ino/g' ajax/ajfile.bak > ajax/ajfile.c perl -pe 's/d_ino/__invalid_d_ino/g' nucleus/embdbi.bak > nucleus/embdbi.c *3. Also you may find the "automake version" problem, try: * aclocal - I m4 autoconf automake -a and then make again. * Best! Zhendong Shaw From ajb at ebi.ac.uk Tue May 2 08:01:52 2006 From: ajb at ebi.ac.uk (ajb at ebi.ac.uk) Date: Tue, 2 May 2006 09:01:52 +0100 (BST) Subject: [EMBOSS] useful hits about install EMBOSS-3.0.0 under cygwin In-Reply-To: References: Message-ID: <34986.81.98.244.247.1146556912.squirrel@webmail.ebi.ac.uk> Thanks for the post. Some quick comments though. > 1. Read the HowTo by Olivier Martin > http://emboss.sourceforge.net/download/cygwin.html Although there is nothing really inaccurate about the HowTo it does imply that --without-x is essential. X windows works fine as long as you've installed it under cygwin. It also suggests using CFLAGS=-s to the configure, presumably to strip the binaries. This should not be required as the latest cygwin distributions automatically strip the binaries. > 2. If you find some compile error about "d_ino"(maybe the most popular > problem ^_^), you could try as follow: > *mv ajax/ajfile.c ajax/ajfile.bak > mv nucleus/embdbi.c nucleus/embdbi.bak > perl -pe 's/d_ino/__invalid_d_ino/g' ajax/ajfile.bak > ajax/ajfile.c > perl -pe 's/d_ino/__invalid_d_ino/g' nucleus/embdbi.bak > > nucleus/embdbi.c This would indeed work but you might as well apply all the fixes from: ftp://emboss.open-bio.org/pub/EMBOSS/fixes/ which fix this and other issues. > *3. Also you may find the "automake version" problem, try: > * aclocal - I m4 > autoconf > automake -a > and then make again. > * I don't think anyone has ever reported "the automake version" problem so I'm unsure what is meant here. HTH Alan From stefan.rensing at biologie.uni-freiburg.de Thu May 4 12:35:22 2006 From: stefan.rensing at biologie.uni-freiburg.de (Stefan Rensing) Date: Thu, 04 May 2006 14:35:22 +0200 Subject: [EMBOSS] sixpack Message-ID: <4459F50A.1080503@biologie.uni-freiburg.de> Dear all, is there a way to create a sixpack output that contains the n.a. ORFs and/or their coordinates in the fasta description line? Also, is there an option to output the longest ORF only? And yes, both would be useful. ;) Cheers, Stefan -- Dr. Stefan Rensing, Group Leader Computational Biology Plant Biotechnology, Faculty of Biology, University of Freiburg Schaenzlestr. 1, D-79104 Freiburg, Fon: +49 761 203-6974, Fax: -6945 http://www.plant-biotech.net/ http://www.cosmoss.org/ stefan.rensing at biologie.uni-freiburg.de "I want you to take him from his hotel, drug him, put him in the front of a car, and run a truck into it at 50 mph." From jison at ebi.ac.uk Thu May 4 14:49:56 2006 From: jison at ebi.ac.uk (Jon Ison) Date: Thu, 4 May 2006 15:49:56 +0100 (BST) Subject: [EMBOSS] sixpack In-Reply-To: <4459F50A.1080503@biologie.uni-freiburg.de> References: <4459F50A.1080503@biologie.uni-freiburg.de> Message-ID: <38707.172.31.70.94.1146754196.squirrel@webmail.ebi.ac.uk> Hi Stefan The short answer is no, but both would be easy to easy to implement as ACD options - perhaps for the next release. Please send an example of the fasta description line you have in mind. Cheers Jon > Dear all, > > is there a way to create a sixpack output that contains the n.a. ORFs > and/or their coordinates in the fasta description line? > > Also, is there an option to output the longest ORF only? > > And yes, both would be useful. ;) > > Cheers, Stefan > > > -- > Dr. Stefan Rensing, Group Leader Computational Biology > Plant Biotechnology, Faculty of Biology, University of Freiburg > Schaenzlestr. 1, D-79104 Freiburg, Fon: +49 761 203-6974, Fax: -6945 > http://www.plant-biotech.net/ http://www.cosmoss.org/ > stefan.rensing at biologie.uni-freiburg.de > > "I want you to take him from his hotel, drug him, put him in the > front of a car, and run a truck into it at 50 mph." > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From d.gatherer at vir.gla.ac.uk Fri May 5 08:41:40 2006 From: d.gatherer at vir.gla.ac.uk (Derek Gatherer) Date: Fri, 05 May 2006 09:41:40 +0100 Subject: [EMBOSS] sixpack Message-ID: <6.2.3.4.1.20060505094057.02ad0ba0@lenzie.gla.ac.uk> Hi Jon et al First of all congrats on the new EMBOSS funding. Stefan has pointed to an important problem - we really need a good _flexible_ translator. The current bunch are powerful and fast but too rigid and in some cases don't appear to do what they say on the tin. See my previous posts: http://emboss.open-bio.org/pipermail/emboss/2006-February/002433.html http://emboss.open-bio.org/pipermail/emboss/2006-February/002434.html I have also some private correspondence with Alan, and Gary Williams, about these issues which I'd be happy to forward with their permission. What we need is an app that enables you to say: take this sequence or sequences, translate them from various x to various y, specifying strand, and give me output format z. Currently we don't quite have that. If you could tweak the existing showseq/prettyseq/sixpack, it would be great. Cheers Derek At 15:49 04/05/2006, you wrote: >Hi Stefan > >The short answer is no, but both would be easy to easy to >implement as ACD options - perhaps for the next release. >Please send an example of the fasta description line you >have in mind. > >Cheers > >Jon > > > > > Dear all, > > > > is there a way to create a sixpack output that contains the n.a. ORFs > > and/or their coordinates in the fasta description line? > > > > Also, is there an option to output the longest ORF only? > > > > And yes, both would be useful. ;) > > > > Cheers, Stefan > > > > > > -- > > Dr. Stefan Rensing, Group Leader Computational Biology > > Plant Biotechnology, Faculty of Biology, University of Freiburg > > Schaenzlestr. 1, D-79104 Freiburg, Fon: +49 761 203-6974, Fax: -6945 > > http://www.plant-biotech.net/ http://www.cosmoss.org/ > > stefan.rensing at biologie.uni-freiburg.de > > > > "I want you to take him from his hotel, drug him, put him in the > > front of a car, and run a truck into it at 50 mph." > > _______________________________________________ > > EMBOSS mailing list > > EMBOSS at lists.open-bio.org > > http://lists.open-bio.org/mailman/listinfo/emboss > > > > >_______________________________________________ >EMBOSS mailing list >EMBOSS at lists.open-bio.org >http://lists.open-bio.org/mailman/listinfo/emboss From Emily.Wong at uts.edu.au Tue May 9 06:23:19 2006 From: Emily.Wong at uts.edu.au (Emily Wong) Date: Tue, 09 May 2006 16:23:19 +1000 Subject: [EMBOSS] emboss on intel macs Message-ID: <44603557.1070302@uts.edu.au> Hi, I'm trying to get emboss running on my intel macbook pro. Will there be a release soon that will allow this? Are there any known workarounds to this problem? Thanks in advance -- Emily Wong Application Support Specialist (Bioinformatics) Computational Research Support Unit University of Technology, Sydney, Australia PO Box 123 Broadway NSW 2007 phone: +61 9514 1723 http://crsu.science.uts.edu.au From ajb at ebi.ac.uk Tue May 9 07:09:58 2006 From: ajb at ebi.ac.uk (ajb at ebi.ac.uk) Date: Tue, 9 May 2006 08:09:58 +0100 (BST) Subject: [EMBOSS] emboss on intel macs In-Reply-To: <44603557.1070302@uts.edu.au> References: <44603557.1070302@uts.edu.au> Message-ID: <33314.172.31.100.160.1147158598.squirrel@webmail.ebi.ac.uk> Hello, As long as you apply all the fixes from: ftp://emboss.open-bio.org/pub/EMBOSS/fixes/ then you should be OK. Check the README there for details. HTH Alan > Hi, > > I'm trying to get emboss running on my intel macbook pro. Will there be > a release soon that will allow this? Are there any known workarounds to > this problem? > > Thanks in advance > > -- > Emily Wong > Application Support Specialist (Bioinformatics) > Computational Research Support Unit > University of Technology, Sydney, Australia > PO Box 123 Broadway NSW 2007 > phone: +61 9514 1723 > http://crsu.science.uts.edu.au > > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From kvddrift at earthlink.net Tue May 9 10:22:29 2006 From: kvddrift at earthlink.net (Koen van der Drift) Date: Tue, 9 May 2006 06:22:29 -0400 Subject: [EMBOSS] emboss on intel macs In-Reply-To: <44603557.1070302@uts.edu.au> References: <44603557.1070302@uts.edu.au> Message-ID: <38B3B778-EF01-4FE4-9D1A-AFD6A76E3B7B@earthlink.net> On May 9, 2006, at 2:23 AM, Emily Wong wrote: > Hi, > > I'm trying to get emboss running on my intel macbook pro. Will > there be > a release soon that will allow this? Are there any known > workarounds to > this problem? You can also use fink to install EMBOSS on your mac. As far as I know there are no issues with intel processors for version 3.0.0. - Koen. From Emily.Wong at uts.edu.au Thu May 11 03:09:29 2006 From: Emily.Wong at uts.edu.au (Emily Wong) Date: Thu, 11 May 2006 13:09:29 +1000 Subject: [EMBOSS] oddcomp result wrong Message-ID: <4462AAE9.40105@uts.edu.au> Hi, Thanks for the information about installing emboss on intel macs earlier. Unfortunately I've run into another problem... My aim is to modify oddcomp to provide some additional functionalities. I have successfully installed emboss from fink. Also I have successfully compiled the source code of oddcomp (obtained from emboss-3.0.0 at emboss). However I get different results to the same input data when I run oddcomp from the fink install compared with running oddcomp from the lone compilation. The results for the lone compile is wrong. My compseq-like input is here: Word size 1 Total count 0 A 5 _Output of fink installed oddcomp_ # # Output from 'oddcomp' # # The Expected frequencies are taken from the file: input_like_COMPSEQ.txt # # Word size: 1 *longsequenceAs* # END # _Output of self compiled oddcomp_ # # Output from 'oddcomp' # # The Expected frequencies are taken from the file: input_like_COMPSEQ.txt # # Word size: 1 * longseqnoAs* # END # Any sugestions as to the cause of the problem? Thanks in advance, Emily From tmargus at ebc.ee Thu May 11 12:21:28 2006 From: tmargus at ebc.ee (=?iso-8859-1?Q?T=F5nu_Margus?=) Date: Thu, 11 May 2006 15:21:28 +0300 Subject: [EMBOSS] Questions about emboss bugfixses, rpm installations and dbxflat In-Reply-To: <33314.172.31.100.160.1147158598.squirrel@webmail.ebi.ac.uk> Message-ID: <000601c674f5$7fdae7e0$901728c1@MULTIVITAMIIN> Hi, How to repair/update rpm installation? I fetched and installed version 3.0.0.2 from http://serine.umdnj.edu/~golharam/biorpms/ Now there is version 3.0.0.3 I don't know are there dbxflat bugs are fixed or not? When I tried to index 16S rRNA sequences (genbank format v9.38 ) fetched from RDP I got an error after ~2h indexing: Database b+tree indexing for flat file databases Processing file ./release9_38_unaligned.gb EMBOSS An error in ajindex.c at line 526: Seek error in ajBtreeCachefetch Command line was dbxflat -dbresource GB -idformat Genbank -directory . -filenames *.gb -dbname SSU9 -release 9.38 -date 11.05.06 -fields id,acc,sv,des,key,org Any help is welcome T?nu Margus -----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: Tuesday, May 09, 2006 10:10 AM To: Emily Wong Cc: emboss at lists.open-bio.org Subject: Re: [EMBOSS] emboss on intel macs Hello, As long as you apply all the fixes from: ftp://emboss.open-bio.org/pub/EMBOSS/fixes/ then you should be OK. Check the README there for details. HTH Alan > Hi, > > I'm trying to get emboss running on my intel macbook pro. Will there be > a release soon that will allow this? Are there any known workarounds to > this problem? > > Thanks in advance > > -- > Emily Wong > Application Support Specialist (Bioinformatics) > Computational Research Support Unit > University of Technology, Sydney, Australia > PO Box 123 Broadway NSW 2007 > phone: +61 9514 1723 > http://crsu.science.uts.edu.au > > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > _______________________________________________ EMBOSS mailing list EMBOSS at lists.open-bio.org http://lists.open-bio.org/mailman/listinfo/emboss From aheyland at ufl.edu Mon May 15 03:15:07 2006 From: aheyland at ufl.edu (Andreas Heyland) Date: Sun, 14 May 2006 20:15:07 -0700 Subject: [EMBOSS] problems installing EMBOSS under Cygwin Message-ID: <4467F23B.7010403@ufl.edu> Hi, I download Emboss package and tried to install it under cygwin. I keep getting the error message cc1.exe - Entry Point not found The procedure entry point_impure_ptr could not be located in the dynamic link library cygwin1.dll when I try to run the: "./configure --without-x CFLAGS=-s" command. I am using: cygwin: 1.5.19-4 gcc: 3.4.4-1 It seems that my cygwin installation might not have worked, i.e. there is an incompatibility issue between GCC and Cygwin. Any suggestions? Thanks a lot Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: aheyland.vcf Type: text/x-vcard Size: 437 bytes Desc: not available URL: From stefan.rensing at biologie.uni-freiburg.de Wed May 17 12:16:35 2006 From: stefan.rensing at biologie.uni-freiburg.de (Stefan Rensing) Date: Wed, 17 May 2006 14:16:35 +0200 Subject: [EMBOSS] strange frames Message-ID: <446B1423.3010202@biologie.uni-freiburg.de> Dear all, apparently, sixpack/transeq handle frames on the reverse complementary strand different than e.g. BLAST does (and different than I would have expected). Using sixpack/transeq to translate this sequence: AGATGACTCCGTTGCAGTTGGCTACTGCTTATATTGCATCAGGGGGTGTGCTTCA Sixpack output: R * L R C S W L L L I L H Q G V C F X F1 D D S V A V G Y C L Y C I R G C A S F2 M T P L Q L A T A Y I A S G G V L F3 1 AGATGACTCCGTTGCAGTTGGCTACTGCTTATATTGCATCAGGGGGTGTGCTTCA 55 ----:----|----:----|----:----|----:----|----:----|----:----| 1 TCTACTGAGGCAACGTCAACCGATGACGAATATAACGTAGTCCCCCACACGAAGT 55 L H S R Q L Q S S S I N C * P T H K F6 X I V G N C N A V A * I A D P P T S * F5 S S E T A T P * Q K Y Q M L P H A E F4 Transeq output: Frame -2: >>test_4 EAHPLMQYKQ*PTATESS Frame -1: >>test_5 *STPPDAI*AVANCNGVIX Frame -3: >>test_6 KHTP*CNISSSQLQRSHL IMHO, it should be: Frame 1 = +1 Frame 2 = +2 Frame 3 = +3 Frame 4 = -1 Frame 5 = -2 Frame 6 = -3 However, in this case, sixpack and transeq return: Frame 1 = +1 Frame 2 = +2 Frame 3 = +3 Frame 4 = -2 Frame 5 = -1 Frame 6 = -3 Is this the expected behaviour, and if so, why? Regards, Stefan -- Dr. Stefan Rensing, Group Leader Computational Biology Plant Biotechnology, Faculty of Biology, University of Freiburg Schaenzlestr. 1, D-79104 Freiburg, Fon: +49 761 203-6974, Fax: -6945 http://www.plant-biotech.net/ http://www.cosmoss.org/ stefan.rensing at biologie.uni-freiburg.de "Mr. Musgrave, please don't interrupt me when I'm asking rhetorical questions." From jison at ebi.ac.uk Wed May 17 13:17:38 2006 From: jison at ebi.ac.uk (Jon Ison) Date: Wed, 17 May 2006 14:17:38 +0100 (BST) Subject: [EMBOSS] problems installing EMBOSS under Cygwin In-Reply-To: <4467F23B.7010403@ufl.edu> References: <4467F23B.7010403@ufl.edu> Message-ID: <46444.172.31.100.168.1147871858.squirrel@webmail.ebi.ac.uk> Hi Andreas I don't think you got a reply to this. I've had a trawl through the messages to the EMBOSS lists ... perhaps some of the following will help (I'm no expert on cygwin) ? A port of EMBOSS 3 to windows will be made available soonish - so you'll never need to use cygwin again. Cheers Jon 1. Apply all the fixes from: ftp://emboss.open-bio.org/pub/EMBOSS/fixes/ 2. Update cygwin Problems can arise from using older versions of cygwin, or selective ("pick and choose") installations of cygwin. Download a new, completely fresh copy of cygwin (the cygwin updating mechanism has proved unsatisfactory) and accept the default packages; best to get a pretty complete set of cygwin packages. If you decide to install a fresh copy then be sure to delete any cygwin registry entries after deleting the cygwin directory: see (1) DELETING CYGWIN below (email emboss-bug if the registry fixing script cygtidy.reg is not available on the website). Your cygwin installation directory should be around 1.5 - 2Gb in size. 3. Read the instructions on the web, but see the corrections (1) to that text, which haven't made it onto the website yet: http://emboss.sourceforge.net/download/cygwin.html The instructions imply that --without-x is essential, whereas X windows works fine as long as you've installed it under cygwin. It also suggests using CFLAGS=-s to the configure, presumably to strip the binaries. This should not be required as the latest cygwin distributions automatically strip the binaries. (1) CORRECTIONS / ADDITIONS 1) install cygwin (http://www.cygwin.com). That's surely possible to compile without it, but it's making everything easier. You'll have to install the base part plus the devel/make and devel/gcc packages. Before installing, see the first line of 2). Then you'll have a linux-like command-line with tools like gcc and make. The paths are then made according to the place where you installed cygwin (ex: c\cygwin becomes /). Notes from the EMBOSS developers: We recommend doing a full cygwin installation or problems may arise with creating the DLLs. 3) quit and restart the cygwin bash (you'll start in your "FirstName_LastName" directory, you can delete "FirstName LastName"), download the EMBOSS package and cd to the place where you uncompressed it. Follow the usual procedure: ./configure --without-x CFLAGS=-s make make install --without-x because at one point the X11 library is required (even if useless) and without this flag the compilation would stop. CFLAGS=-s to make the binaries smaller. Notes from the EMBOSS developers: If you have done a reasonably full cygwin installation then you can leave out the --without-x switch. X will work perfectly well if installed. Also, with later cygwin installations, you can omit the CFLAGS=-s switch. We also recommend providing a prefix switch to install EMBOSS in a specified location. The configuration then becomes: ./configure --prefix=/usr/local/emboss make make install After installation make sure that /usr/local/emboss/bin is in your PATH. DELETING CYGWIN To delete cygwin you should drag your cygwin installation directory to the recycle bin. Then delete any cygwin entries from the 'start' menu. You should then download the file cygtidy.reg and double-click it to remove any cygwin registry entries. > Hi, I download Emboss package and tried to install it under cygwin. I > keep getting the error message cc1.exe - Entry Point not found The > procedure entry point_impure_ptr could not be located in the dynamic > link library cygwin1.dll when I try to run the: "./configure --without-x > CFLAGS=-s" command. I am using: > cygwin: 1.5.19-4 > gcc: 3.4.4-1 > It seems that my cygwin installation might not have worked, i.e. there > is an incompatibility issue between GCC and Cygwin. Any suggestions? > Thanks a lot > Andreas > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From sbassi at gmail.com Wed May 17 15:04:15 2006 From: sbassi at gmail.com (Sebastian Bassi) Date: Wed, 17 May 2006 12:04:15 -0300 Subject: [EMBOSS] problems installing EMBOSS under Cygwin In-Reply-To: <46444.172.31.100.168.1147871858.squirrel@webmail.ebi.ac.uk> References: <4467F23B.7010403@ufl.edu> <46444.172.31.100.168.1147871858.squirrel@webmail.ebi.ac.uk> Message-ID: On 5/17/06, Jon Ison wrote: > I don't think you got a reply to this. I've had a trawl through the > messages to the EMBOSS lists ... perhaps some of the following will > help (I'm no expert on cygwin) ? A port of EMBOSS 3 to windows will > be made available soonish - so you'll never need to use cygwin again. There is another solution: Use EMBOSS w/o install Linux, with a LiveCD. That is why I included EMBOSS 3.0 in DNALinux Desktop edition: http://www.dnalinux.com It is free as a GPL :) -- Bioinformatics news: http://www.bioinformatica.info Lriser: http://www.linspire.com/lraiser_success.php?serial=318 From Staffan.X.Bergh at astrazeneca.com Thu May 18 10:48:31 2006 From: Staffan.X.Bergh at astrazeneca.com (Staffan.X.Bergh at astrazeneca.com) Date: Thu, 18 May 2006 12:48:31 +0200 Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Message-ID: Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box with RedHat Enterprise Server 3 (not the latest, I know, but corporate standards, etc, etc, ...) The error message from make/ld is: > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when searching for -lX11 > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when searching for -lX11 > /usr/bin/ld: cannot find -lX11 The X version that RHES3 ships with is 4.3.0 > $ X -version > XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: 4.3.0-68.EL) > Release Date: 15 August 2003 > (and a lot more, deleted by the RSSPBW [Royal Swedish Society for the Preservation of BandWidth]). I couldn't find answer to this in the docs, nor in the mail archives, and I guess that I'll end up having to upgrade X, but it would be nice if someone could confirm that that would solve the problem before I try? Or is this something else? cheers /staffan _____________________________________________________________ Staffan Bergh, PhD Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D B209:537, SE-151 81 S?dert?lje, Sweden Tel: +46 8 553 215 72 staffan.x.bergh at astrazeneca.com From Marc.Logghe at DEVGEN.com Thu May 18 12:30:00 2006 From: Marc.Logghe at DEVGEN.com (Marc Logghe) Date: Thu, 18 May 2006 14:30:00 +0200 Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Message-ID: <0C528E3670D8CE4B8E013F6749231AA6746DD0@ANTARESIA.be.devgen.com> I ran into the same issue a while ago. If I recall well you had to run configure with the option '--without-x'. HTH, Marc > -----Original Message----- > From: emboss-bounces at lists.open-bio.org > [mailto:emboss-bounces at lists.open-bio.org] On Behalf Of > Staffan.X.Bergh at astrazeneca.com > Sent: Thursday, May 18, 2006 12:49 PM > To: emboss at emboss.open-bio.org > Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? > > Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box > with RedHat Enterprise Server 3 (not the latest, I know, but > corporate standards, etc, etc, ...) > > The error message from make/ld is: > > > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when > > searching > for -lX11 > > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when > > searching > for -lX11 > > /usr/bin/ld: cannot find -lX11 > > The X version that RHES3 ships with is 4.3.0 > > > $ X -version > > XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: > > 4.3.0-68.EL) Release Date: 15 August 2003 (and a lot more, > deleted by > > the RSSPBW [Royal Swedish Society for the > Preservation of BandWidth]). > > I couldn't find answer to this in the docs, nor in the mail > archives, and I guess that I'll end up having to upgrade X, > but it would be nice if someone could confirm that that would > solve the problem before I try? > Or is this something else? > > cheers /staffan > _____________________________________________________________ > Staffan Bergh, PhD > Informatics and Logistics, Molecular Pharmacology, > AstraZeneca R&D B209:537, SE-151 81 S?dert?lje, Sweden > Tel: +46 8 553 215 72 > staffan.x.bergh at astrazeneca.com > > > _______________________________________________ > EMBOSS mailing list > EMBOSS at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/emboss > From ajb at ebi.ac.uk Thu May 18 14:21:24 2006 From: ajb at ebi.ac.uk (ajb at ebi.ac.uk) Date: Thu, 18 May 2006 15:21:24 +0100 (BST) Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? In-Reply-To: References: Message-ID: <47180.81.98.244.247.1147962084.squirrel@webmail.ebi.ac.uk> If you do a: setenv CFLAGS "-D__amd64__" (4 underscores in all!) before configuring then that will stop the segfaulting on opterons. You should at least 'make clean' before configuring again. The above will not be required in the next release of EMBOSS. Alan > Hi again, > > the machine is an HP DL385 med 2xOpteron 2.8GHz CPUs ... > and yes, I've checked for the 64bit libraries, too. > > To add to my woes, configuring "--without-x" (thanks Mark > for the pointer!) and compiling (w/o warnings but for two > incompatible pointers in ajfmt.c) gives binaries that segfault > immediately. The config.log has some more > > '/usr/bin/ld: skipping incompatible' library, > > but now for libz, libm and libc in the conftest for inflateEnd > in -lz ... So I tried turning configuring w/o the pngdriver, but > still get segfault'ing binaries. > > So I'm at a bit of a loss about what to do right > now - except force our IS/IT dept to upgrade RedHat ... > > If anyone want to see the output from configure and/or make you're > welcome (but no sniffing around ;-) > > cheers /staffan > > -----Original Message----- > From: ajb at ebi.ac.uk [mailto:ajb at ebi.ac.uk] > Sent: 18 May 2006 15:17 > To: Staffan.X.Bergh at astrazeneca.com > Subject: RE: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? > > > Hi Staffan, > > Oh dear. Is it an AMD64 or EM64T by any chance as that error might make > sense > if it was trying to combine 32bit and 64 bit libraries. Noone has > reported it before though - but I expect few of our users have > RHES. If it is a 64bit m/c you will, of course, have checked that > both 32bit and 64bit libs/devel files have been installed. > > Also, as you're corporate, there is patently no chance of a sniff > around from this end. > > It looks like an X11 upgrade might be worth a try. > > Cheers > Alan > > > > > > > > >> Hi Alan, >> >> I do have the devel package installed ... >> >>> $ rpm -q -f /usr/X11R6/lib/libX11.a -i >>> Name : XFree86-devel Relocations: (not >>> relocatable) >>> Version : 4.3.0 Vendor: Red Hat, Inc. >>> Release : 68.EL Build Date: Sun 22 Aug 2004 >> 04:22:47 PM CEST >>> Install Date: Sun 25 Sep 2005 01:05:22 AM CEST Build Host: >> tweety.build.redhat.com >>> Group : Development/Libraries Source RPM: >> XFree86-4.3.0-68.EL.src.rpm >>> Size : 13528038 License: >> MIT/X11/XFree86/many others >>> Signature : DSA/SHA1, Mon 23 Aug 2004 07:58:10 PM CEST, Key ID >> 219180cddb42a60e >>> Packager : Red Hat, Inc. >>> URL : http://www.xfree86.org >>> Summary : X11R6 static libraries, headers and programming man >>> pages. >>> Description : >>> XFree86-devel includes the libraries, header files and documentation >>> needed for developing applications for the X window system. >> >> and it verifies OK, so it's not broken in some mysterious way ... >> >> cheers /staffan >> >> -----Original Message----- >> From: ajb at ebi.ac.uk [mailto:ajb at ebi.ac.uk] >> Sent: 18 May 2006 13:26 >> To: Staffan.X.Bergh at astrazeneca.com >> Cc: emboss-bug at emboss.open-bio.org >> Subject: Re: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? >> >> >> Hi Staffan, >> >> First check that you have the XFree86-devel RPM installed. >> If not, then install it, do a make clean and then configure >> again. >> >> Let us know whether that solves it. I've never seen that error >> in quite that form before but the XFree86-devel RPM is essential. >> >> Alan Bleasby >> EBI >> >> PS: CCing to the rest of the developers >> >> >>> Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box with >>> RedHat >>> Enterprise Server 3 (not >>> the latest, I know, but corporate standards, etc, etc, ...) >>> >>> The error message from make/ld is: >>> >>>> /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when >>>> searching >>> for -lX11 >>>> /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when >>>> searching >>> for -lX11 >>>> /usr/bin/ld: cannot find -lX11 >>> >>> The X version that RHES3 ships with is 4.3.0 >>> >>>> $ X -version >>>> XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: >>>> 4.3.0-68.EL) >>>> Release Date: 15 August 2003 >>>> (and a lot more, deleted by the RSSPBW [Royal Swedish Society for the >>> Preservation of BandWidth]). >>> >>> I couldn't find answer to this in the docs, nor in the mail archives, >>> and >>> I >>> guess that I'll end up having to >>> upgrade X, but it would be nice if someone could confirm that that >>> would >>> solve the problem before I try? >>> Or is this something else? >>> >>> cheers /staffan >>> _____________________________________________________________ >>> Staffan Bergh, PhD >>> Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D >>> B209:537, SE-151 81 S?dert?lje, Sweden >>> Tel: +46 8 553 215 72 >>> staffan.x.bergh at astrazeneca.com >>> >>> >>> _______________________________________________ >>> EMBOSS mailing list >>> EMBOSS at lists.open-bio.org >>> http://lists.open-bio.org/mailman/listinfo/emboss >>> >>> >> >> >> > > From Staffan.X.Bergh at astrazeneca.com Thu May 18 14:11:13 2006 From: Staffan.X.Bergh at astrazeneca.com (Staffan.X.Bergh at astrazeneca.com) Date: Thu, 18 May 2006 16:11:13 +0200 Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Message-ID: Hi again, the machine is an HP DL385 med 2xOpteron 2.8GHz CPUs ... and yes, I've checked for the 64bit libraries, too. To add to my woes, configuring "--without-x" (thanks Mark for the pointer!) and compiling (w/o warnings but for two incompatible pointers in ajfmt.c) gives binaries that segfault immediately. The config.log has some more '/usr/bin/ld: skipping incompatible' library, but now for libz, libm and libc in the conftest for inflateEnd in -lz ... So I tried turning configuring w/o the pngdriver, but still get segfault'ing binaries. So I'm at a bit of a loss about what to do right now - except force our IS/IT dept to upgrade RedHat ... If anyone want to see the output from configure and/or make you're welcome (but no sniffing around ;-) cheers /staffan -----Original Message----- From: ajb at ebi.ac.uk [mailto:ajb at ebi.ac.uk] Sent: 18 May 2006 15:17 To: Staffan.X.Bergh at astrazeneca.com Subject: RE: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Hi Staffan, Oh dear. Is it an AMD64 or EM64T by any chance as that error might make sense if it was trying to combine 32bit and 64 bit libraries. Noone has reported it before though - but I expect few of our users have RHES. If it is a 64bit m/c you will, of course, have checked that both 32bit and 64bit libs/devel files have been installed. Also, as you're corporate, there is patently no chance of a sniff around from this end. It looks like an X11 upgrade might be worth a try. Cheers Alan > Hi Alan, > > I do have the devel package installed ... > >> $ rpm -q -f /usr/X11R6/lib/libX11.a -i >> Name : XFree86-devel Relocations: (not >> relocatable) >> Version : 4.3.0 Vendor: Red Hat, Inc. >> Release : 68.EL Build Date: Sun 22 Aug 2004 > 04:22:47 PM CEST >> Install Date: Sun 25 Sep 2005 01:05:22 AM CEST Build Host: > tweety.build.redhat.com >> Group : Development/Libraries Source RPM: > XFree86-4.3.0-68.EL.src.rpm >> Size : 13528038 License: > MIT/X11/XFree86/many others >> Signature : DSA/SHA1, Mon 23 Aug 2004 07:58:10 PM CEST, Key ID > 219180cddb42a60e >> Packager : Red Hat, Inc. >> URL : http://www.xfree86.org >> Summary : X11R6 static libraries, headers and programming man pages. >> Description : >> XFree86-devel includes the libraries, header files and documentation >> needed for developing applications for the X window system. > > and it verifies OK, so it's not broken in some mysterious way ... > > cheers /staffan > > -----Original Message----- > From: ajb at ebi.ac.uk [mailto:ajb at ebi.ac.uk] > Sent: 18 May 2006 13:26 > To: Staffan.X.Bergh at astrazeneca.com > Cc: emboss-bug at emboss.open-bio.org > Subject: Re: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? > > > Hi Staffan, > > First check that you have the XFree86-devel RPM installed. > If not, then install it, do a make clean and then configure > again. > > Let us know whether that solves it. I've never seen that error > in quite that form before but the XFree86-devel RPM is essential. > > Alan Bleasby > EBI > > PS: CCing to the rest of the developers > > >> Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box with >> RedHat >> Enterprise Server 3 (not >> the latest, I know, but corporate standards, etc, etc, ...) >> >> The error message from make/ld is: >> >>> /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when >>> searching >> for -lX11 >>> /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when >>> searching >> for -lX11 >>> /usr/bin/ld: cannot find -lX11 >> >> The X version that RHES3 ships with is 4.3.0 >> >>> $ X -version >>> XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: 4.3.0-68.EL) >>> Release Date: 15 August 2003 >>> (and a lot more, deleted by the RSSPBW [Royal Swedish Society for the >> Preservation of BandWidth]). >> >> I couldn't find answer to this in the docs, nor in the mail archives, >> and >> I >> guess that I'll end up having to >> upgrade X, but it would be nice if someone could confirm that that would >> solve the problem before I try? >> Or is this something else? >> >> cheers /staffan >> _____________________________________________________________ >> Staffan Bergh, PhD >> Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D >> B209:537, SE-151 81 S?dert?lje, Sweden >> Tel: +46 8 553 215 72 >> staffan.x.bergh at astrazeneca.com >> >> >> _______________________________________________ >> EMBOSS mailing list >> EMBOSS at lists.open-bio.org >> http://lists.open-bio.org/mailman/listinfo/emboss >> >> > > > From golharam at umdnj.edu Thu May 18 16:00:29 2006 From: golharam at umdnj.edu (Ryan Golhar) Date: Thu, 18 May 2006 12:00:29 -0400 Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? In-Reply-To: Message-ID: <00a501c67a94$2f7ac000$2f01a8c0@GOLHARMOBILE1> That's interesting. We are running Enterprise v3 here, both workstation and server and I haven't seen this. We are also running with the latest patches from RedHat. We've also built RPMs specifically for RHEL v3 for EMBOSS (and other applications). You can get them from here: http://serine.umdnj.edu/~golharam/biorpms Ryan -----Original Message----- From: emboss-bounces at lists.open-bio.org [mailto:emboss-bounces at lists.open-bio.org] On Behalf Of Staffan.X.Bergh at astrazeneca.com Sent: Thursday, May 18, 2006 6:49 AM To: emboss at emboss.open-bio.org Subject: [EMBOSS] X11 version requirement for EMBOSS 3.0.0? Hi, I'm trying to compile/install EMBOSS 3.0.0 on a linux box with RedHat Enterprise Server 3 (not the latest, I know, but corporate standards, etc, etc, ...) The error message from make/ld is: > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.so when > searching for -lX11 > /usr/bin/ld: skipping incompatible /usr/X11R6/lib/libX11.a when > searching for -lX11 > /usr/bin/ld: cannot find -lX11 The X version that RHES3 ships with is 4.3.0 > $ X -version > XFree86 Version 4.3.0 (Red Hat Enterprise Linux 3 release: 4.3.0-68.EL) > Release Date: 15 August 2003 > (and a lot more, deleted by the RSSPBW [Royal Swedish Society for the Preservation of BandWidth]). I couldn't find answer to this in the docs, nor in the mail archives, and I guess that I'll end up having to upgrade X, but it would be nice if someone could confirm that that would solve the problem before I try? Or is this something else? cheers /staffan _____________________________________________________________ Staffan Bergh, PhD Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D B209:537, SE-151 81 S?dert?lje, Sweden Tel: +46 8 553 215 72 staffan.x.bergh at astrazeneca.com _______________________________________________ EMBOSS mailing list EMBOSS at lists.open-bio.org http://lists.open-bio.org/mailman/listinfo/emboss From Staffan.X.Bergh at astrazeneca.com Fri May 19 13:47:10 2006 From: Staffan.X.Bergh at astrazeneca.com (Staffan.X.Bergh at astrazeneca.com) Date: Fri, 19 May 2006 15:47:10 +0200 Subject: [EMBOSS] Solved! Was: X11 version requirement for EMBOSS 3.0.0? Message-ID: OK, got the X problem worked around now. Thanks all who answered (and special thanks to Alan for not blasting my ears off for missing the FAQ about the Opterons ...) In summary: On our bright new box with 2xOpteron processors, and RedHat Enterprise Server 3.0, I was trying to compile EMBOSS-3.0.0. I got strange errors from the linker, like skipping incompatible "/usr/X11R6/lib/libX11.a" Solution was to upgrade the X libraries: version 4.6.0 Xprg.tgz file from the Xfree86.org ftp site, install in a convenient location and then point configure at the new X libraries: ./configure --x-libraries=/usr/local/X11/lib64/ \ --x-includes=/usr/local/X11/include/ That solved the compilation problem. Setting Pseudocolor in Exceed then solved next (and last, I hope) problem: Error in XCreatePixmap: BadDrawable (invalid Pixmap or Window parameter) The actual reason for the linking problem remains unclear, but I note that the rpm we had installed (that came with RHEL) had both static and shared 32bit libraries, but only shared 64bit libraries ... so maybe something not quite standard happened at RedHat when they made the rpm ... cheers /staffan _____________________________________________________________ Staffan Bergh, PhD Informatics and Logistics, Molecular Pharmacology, AstraZeneca R&D B209:537, SE-151 81 S?dert?lje, Sweden Tel: +46 8 553 215 72 staffan.x.bergh at astrazeneca.com