From keith.jolley at medawar.oxford.ac.uk Mon Jan 5 10:46:59 2004 From: keith.jolley at medawar.oxford.ac.uk (Keith Jolley) Date: Mon, 5 Jan 2004 15:46:59 -0000 Subject: [Bio-Linux] Broken desktop links using redhat-config-users to create new user accounts Message-ID: I've run into a small problem with creating new user accounts using /usr/bin/redhat-config-users. The desktop symlink to 'Bioinformatics software manuals' is not replicated from /etc/skel properly, leading to a broken link. Listing the contents of the new user's .gnome-desktop directory shows that there is junk data added to the symlink: lrwxrwxrwx 1 keith keith 46 Jan 5 15:07 Bioinformatics software manuals -> /usr/software/documentation? ?? ?? Hello all, I just finished reading about and trying out the program Pfaat, which is a graphical protein multiple sequence alignment program, but with lots of bells and whistles. Information about the program, and downloads can be found at: http://pfaat.sourceforge.net/ Ripping info straight off their page: "Key features of Pfaat include the ability to align collections of sequences, cluster and/or group sequences into subfamilies, analyze sequences based on a number of similarity criteria, visualize protein structure, and annotate sequences and specific residue positions with text descriptions. " Dan and I just played around with it and were both quite impressed. We are considering putting this onto a future release of Bio-Linux, and if any of you decide to give it a try, please let us know what you think so we know whether it is as useful to the user community as we think it could be. If anyone does try this out - we'd be interested to know if the SRS functionality works for you or not. We ran the program on a machine with SRS already installed, so I'm not sure whether how they have implemented the SRS functions - local or via the web. (I know I could dig around for this and figure it out, but maybe one of you kind souls will let us know your experiences should you try it.) Installation: All it took for me to install this on Bio-Linux was the following: Make a directory for the installation and put the downloaded file into this directory Run the command: sh install.exe (or similar - check the instructions for the program!) Make a subdirectory from wherever you put the distribution called "bin", then make a subdirectory under bin called "linux" Inside that directory (bin/linux), make softlinks to clustalw, hmmbuild, hmmalign: ln -s /usr/software/exec/clustalw clustalw ln -s /usr/software/exec/hmmbuild hmmbuild ln -s /usr/software/exec/hmmalign hmmalign I also set the environmental variable PFAAT to point to the directrory where the distribution was. (I put this in my home directory .zshrc file, but if you wanted it to be universally applicable, you could put it in /usr/software/bioenvrc, or into /etc/zshrc...or whatever the appropriate global file is for the default shell on your machine if you have changed it.) As I recall, that's all I did. They have an example directory where you can load up an alignment, etc. Please let us know if you try this program and what you think. We're also always happy to hear about software you discover out there that looks like it might be of interest to other Bio-Linux users. cheers Bela ************************* Dr. Bela Tiwari Lead Bioinformatician CEH Oxford Mansfield Road Oxford, OX1 3SR 01865 281975 ************************* From nsb at ceh.ac.uk Thu Jan 8 13:40:53 2004 From: nsb at ceh.ac.uk (Nicolas Bertrand) Date: Thu, 08 Jan 2004 18:40:53 +0000 Subject: [Bio-Linux] Mult. sequence alignment editor program of interest forBio-Linux users? Message-ID: Really nice. Thank you! Tried it on Windows and Linux. Problems on Bio-Linux (and Windows): pfaat could not find hmmbuild (even so hmmbuild is in the path in Linux, on windows hmmbuild was not even installed so no surprise, there :-). Would be interesting to hear if somebody has got this working. SRS Searching works. Should you want to query a different SRS server (or change the database or field you need to search!), you have to edit the file pfaat.properties in the pfaat installation directory the default line is: pfaat.srsServer=srs.ebi.ac.uk/srs5bin/cgi-bin/wgetz?-e+[swall-id:QUERY_NAME]+-vn+2 I have changed it to point out to our internal SRS server as follows: "pfaat.srsServer=ivgfs.nox.ac.uk/srs71bin/cgi-bin/wgetz?-e+[swall-id:QUERY_NAME]+-vn+2" and could get some results back (See attachment). The tree viewing functionality seems taken directly from ATV, software that is already included in Bio-Linux :-)[http://www.genetics.wustl.edu/eddy/atv/] Nic ********************************************** Nicolas Bertrand IT Support, GRID & Bioinformatics CEH Oxford, Mansfield Road Oxford OX1 3SR ********************************************** PS: I have to say though that pfaat feels significantly snappier on a windows box than on Linux :-( >>> "Bela Tiwari" 08/01/2004 14:44:39 >>> Hello all, I just finished reading about and trying out the program Pfaat, which is a graphical protein multiple sequence alignment program, but with lots of bells and whistles. Information about the program, and downloads can be found at: http://pfaat.sourceforge.net/ Ripping info straight off their page: "Key features of Pfaat include the ability to align collections of sequences, cluster and/or group sequences into subfamilies, analyze sequences based on a number of similarity criteria, visualize protein structure, and annotate sequences and specific residue positions with text descriptions. " Dan and I just played around with it and were both quite impressed. We are considering putting this onto a future release of Bio-Linux, and if any of you decide to give it a try, please let us know what you think so we know whether it is as useful to the user community as we think it could be. If anyone does try this out - we'd be interested to know if the SRS functionality works for you or not. We ran the program on a machine with SRS already installed, so I'm not sure whether how they have implemented the SRS functions - local or via the web. (I know I could dig around for this and figure it out, but maybe one of you kind souls will let us know your experiences should you try it.) Installation: All it took for me to install this on Bio-Linux was the following: Make a directory for the installation and put the downloaded file into this directory Run the command: sh install.exe (or similar - check the instructions for the program!) Make a subdirectory from wherever you put the distribution called "bin", then make a subdirectory under bin called "linux" Inside that directory (bin/linux), make softlinks to clustalw, hmmbuild, hmmalign: ln -s /usr/software/exec/clustalw clustalw ln -s /usr/software/exec/hmmbuild hmmbuild ln -s /usr/software/exec/hmmalign hmmalign I also set the environmental variable PFAAT to point to the directrory where the distribution was. (I put this in my home directory .zshrc file, but if you wanted it to be universally applicable, you could put it in /usr/software/bioenvrc, or into /etc/zshrc...or whatever the appropriate global file is for the default shell on your machine if you have changed it.) As I recall, that's all I did. They have an example directory where you can load up an alignment, etc. Please let us know if you try this program and what you think. We're also always happy to hear about software you discover out there that looks like it might be of interest to other Bio-Linux users. cheers Bela ************************* Dr. Bela Tiwari Lead Bioinformatician CEH Oxford Mansfield Road Oxford, OX1 3SR 01865 281975 ************************* _______________________________________________ Bio-Linux mailing list Bio-Linux at ivsun01.nerc-oxford.ac.uk http://www.bioinf.ceh.ac.uk/mailman/listinfo/bio-linux -------------- next part -------------- A non-text attachment was scrubbed... Name: srs_pfaat.JPG Type: image/jpeg Size: 56552 bytes Desc: not available URL: From dswan at ceh.ac.uk Mon Jan 12 05:09:38 2004 From: dswan at ceh.ac.uk (Dan Swan) Date: Mon, 12 Jan 2004 10:09:38 +0000 Subject: [Bio-Linux] Bio-Linux 3.0 installations and updates. Message-ID: <40027262.70002@ceh.ac.uk> Hi All, The initial rush of Bio-Linux 3.0 installations has slowed, so we currently have a number of slots available for those of you who have not yet upgraded. We offer 2 methods of updating - a complete reinstall for people who just prefer to back up their data and make a clean start, or an upgrade which preserves your data, but is slightly more involved. This is all covered in the following document: http://envgen.nox.ac.uk/envgen/software/archives/000432.html Bio-Linux 1.0/2.0/2.1 users who have not upgraded are advised to do so soon as ReHat 7.3 support (the platform on with versions 1 - 2.1 are based) is shortly being withdrawn. If you installed Bio-Linux 3.0 for beta testing purposes prior to Christmas you may wish to synchronise your machine against our latest image. Please contact the helpdesk (helpdesk at envgen.nox.ac.uk) for details. Recent updates to the Bio-Linux image include: Latest kernel (2.4.20-28.9) Updated rsync (rsync 2.5.7-0.9) Updated Apache (2.0.40-21.9) plus many other updates. There have also been some fixes applied to the image thanks to bug supports reported by various people, thanks to everyone who contributed. regards, Dan and the Bio-Linux team -- Dr Dan Swan (RHCE) - Bio-Linux Developer EGTDC, CEH, Mansfield Road, Oxford, OX1 3SR Tel: 01865 281 658 Fax: 01865 281 696 http://bio-linux.net/ | dswan at ceh.ac.uk From dswan at ceh.ac.uk Wed Jan 14 04:48:08 2004 From: dswan at ceh.ac.uk (Dan Swan) Date: Wed, 14 Jan 2004 09:48:08 +0000 Subject: [Bio-Linux] Broken File Associations in Nautilus Message-ID: <40051058.2050809@ceh.ac.uk> Hi All, It has come to my attention (thanks Sarah!) that if you use the file manager "Nautilus" on Bio-Linux 3.0 certain behaviours do not work as expected. Specifically it is impossible to "double click" OpenOffice files, Excel files and Word files and automatically launch OpenOffice. In order to correctly associate these file types with OpenOffice you can do the following (please note these instructions can be applied to any program/document you wish to autoload): 1) Make sure you have run OpenOffice at least once and gone through the setup. 2) Open Nautilus and navigate to a file you cannot double-click. 3) Right click on the file. Select "Open with" and "Other Application". Note - this works best on icon not list view. 4) A box should pop up. You can then clcik the "Go There" button next to "File Types and Programs". 5) A new window appears "Edit File type". 6) Default action should read "OpenOffice.org" - change this to "Custom". 7) Then where it says "Program to Run" click "Browse". 8) From your home directory double click "OpenOffice.org1.1.0" From the right hand pane select "soffice" and click "OK". Click "OK" again to close the "Edit file types". 9)Double click on any file of that type and it will now open in OpenOffice. regards, Dan -- Dr Dan Swan (RHCE) - Bio-Linux Developer EGTDC, CEH, Mansfield Road, Oxford, OX1 3SR Tel: 01865 281 658 Fax: 01865 281 696 http://bio-linux.net/ | dswan at ceh.ac.uk From anjw at ceh.ac.uk Wed Jan 14 09:26:53 2004 From: anjw at ceh.ac.uk (anjw) Date: 14 Jan 2004 14:26:53 +0000 Subject: [Bio-Linux] Advanced Bio-Linux Course Message-ID: <1074090412.18309.19.camel@ivpcp030.nox.ac.uk> Dear Colleagues A few places remain available on the Advanced Bio-Linux course to be held in Oxford on January 26th. If you have yet to attend this practical-based course or wish to attend again see the following website for details: http://envgen.nox.ac.uk/training.html To register for this course, please email the EGTDC helpdesk - helpdesk at envgen.nox.ac.uk. Best Wishes The EGTDC team From dswan at ceh.ac.uk Tue Jan 27 10:09:17 2004 From: dswan at ceh.ac.uk (Dan Swan) Date: Tue, 27 Jan 2004 15:09:17 +0000 Subject: [Bio-Linux] Bio-Linux update Message-ID: <40167F1D.8050102@ceh.ac.uk> Hello all, This is just a reminder to you all that Red Hat 7.3, which Bio-Linux 1.X and 2.X were based on is no longer being updated with security patches. Consequently this is another call for people to take up upgrade slots as we phase out support for versions of Bio-Linux prior to 3.0. Those of you on non-EG grants are more than welcome to join this upgrade path as we are very grateful for all the input you have given us into the project as early external adopters. The Bio-Linux team is currently working on ways to provide Bio-Linux more widely available. The project now has space at bioinformatics.org the biggest resource for Open Source bioinformatics. We may migrate the mailing lists there at some point, but the upshot of this is that colleagues external to the programme who wish to get Bio-Linux will be able to do so from this site. The site also supports a bug reporting system that we might start using for reporting issues. We will soon be issuing a new SOP for Bio-Linux installs external to the EG community. We are also considering major changes to the underlying OS of Bio-Linux for future versions. This has come about by Red Hat's migration towards the enterprise and the spin off of Fedora. This means we are currently in the process of evaluating other Linux distributions in order to give Bio-Linux the most supported, most stable base we can. Currently our list of candidates is SuSE, Fedora (Red Hat), Debian, Knoppix (Debian), Morphix and BioBrew (Red Hat). If anyone has reasonable points they would like to make for, or against these choices, then we would be very interested to hear them. There have been minor updates to Bio-Linx 3.0 since the last mailing. Your Yum logs (/var/log/yum.log*) should include: gaim 1:0.75-0.9.0.i386 mc 1:4.6.0-7.9.i386 slocate 2.7-2.i386 cvs 1.11.2-13.i386 kdepim-devel 6:3.1-6.i386 tcpdump 14:3.7.2-7.9.1.i386 kdepim 6:3.1-6.i386 libpcap 14:0.7.2-7.9.1.i386 ethereal 0.10.0a-0.90.1.i386 ethereal-gnome 0.10.0a-0.90.1.i386 regards, Dan -- Dr Dan Swan - Bio-Linux Developer | RHCE EGTDC, CEH, Mansfield Road, Oxford, OX1 3SR Tel: 01865 281 658 Fax: 01865 281 696 http://bio-linux.net/ | dswan at ceh.ac.uk From dswan at ceh.ac.uk Wed Jan 28 03:44:31 2004 From: dswan at ceh.ac.uk (Dan Swan) Date: Wed, 28 Jan 2004 08:44:31 +0000 Subject: [Bio-Linux] Firewall logs and backup drives (maximal mount count reached errors) Message-ID: <4017766F.3040103@ceh.ac.uk> Dear all, It has been pointed out that currently there is no /var/log/firewall.log for firewall alerts as there was in Bio-Linux 2.X. These messages are currently being logged but not in a centralised location. To remedy this please add the following line to /etc/syslog.conf at the top of the file: kern.=debug -/var/log/firewall.log then restart syslogd with: sudo pkill -HUP syslogd This is now fixed on the latest image. Also if you are seeing messages like this: Jan 26 04:02:01 yeastimp kernel: EXT3-fs warning: maximal mount count reached, running e2fsck is recommended in /var/log/messages then it's because the backup drive /dev/hdb1 has been mounted by the backup script enough times to trigger a warning limit. It is not a major problem. In order to comply with the suggestion you can run the following: sudo umount /backups sudo /sbin/e2fsck /dev/hdb1 You could add these lines to a file called check.sh in /etc/cron.monthly to ensure a monthly check takes place. #!/bin/sh umount /backups /sbin/e2fsck /dev/hdb1 regards, Dan -- Dr Dan Swan - Bio-Linux Developer | RHCE EGTDC, CEH, Mansfield Road, Oxford, OX1 3SR Tel: 01865 281 658 Fax: 01865 281 696 http://bio-linux.net/ | dswan at ceh.ac.uk From dswan at ceh.ac.uk Wed Jan 28 07:22:30 2004 From: dswan at ceh.ac.uk (Dan Swan) Date: Wed, 28 Jan 2004 12:22:30 +0000 Subject: [Bio-Linux] Firewall logs and backup drives "2" Message-ID: <4017A986.8090905@ceh.ac.uk> Just to clarify my previous email: Those of you who have edited syslog.conf (this is the master configuration file the system event logger) will notice that the line I gave to add in is slightly different to those already in there. I suggested: kern.=debug -/var/log/firewall.log if you wish to maintain the look and feel of syslog conf: kern.=debug /var/log/firewall.log is equally appropriate - removing the trailing "-" is not an issue. I also forgot to mention (thanks to Rob Ansderson of Imperial College) that if you were to create a cron.monthly job to e2fsck the backup drive (basically think of it as a disk integrity checker like Windows Scandisk) you will need to chmod 755 the file you create: sudo chmod 755 /etc/cron.monthly/check.sh otherwise it will not run! Cheeers, Dan -- Dr Dan Swan - Bio-Linux Developer | RHCE EGTDC, CEH, Mansfield Road, Oxford, OX1 3SR Tel: 01865 281 658 Fax: 01865 281 696 http://bio-linux.net/ | dswan at ceh.ac.uk