From edward.kawas at gmail.com Sun Oct 3 00:09:33 2004 From: edward.kawas at gmail.com (Eddie Kawas) Date: Sun Oct 3 00:10:55 2004 Subject: [MOBY-dev] Tools for creating service types, service instances, and objects In-Reply-To: <4155C12D.6020305@illuminae.com> References: <4155C12D.6020305@illuminae.com> Message-ID: Hi, I have created some new tools to aid in the construction of service types, service instances, and objects. This is the first release of the tools and I was hoping to see what people think. Anyways, the url for: *objects http://mobycentral.cbr.nrc.ca/applets/object.html *service instances http://mobycentral.cbr.nrc.ca/applets/service.html *service types http://mobycentral.cbr.nrc.ca/applets/servicetype.html The tools are java applets and are 'live'. The initial download is quite large because of some libraries that are needed. Eddie From gss at ncgr.org Wed Oct 6 16:14:26 2004 From: gss at ncgr.org (Gary Schiltz) Date: Wed Oct 6 16:17:34 2004 Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: <415440E3.2070000@ncgr.org> References: <415440E3.2070000@ncgr.org> Message-ID: <41645222.1060907@ncgr.org> I've put information about the MOBY meeting on the Semantic MOBY web site - the URL is www.semanticmoby.org/meeting. It is currently missing a registration form, which I'll add by the end of the day. // Gary Gary Schiltz wrote: > Hello MOBY'ers! > > The next MOBY meeting will be held at NCGR (www.ncgr.org) in Santa Fe, > New Mexico, the weekend of 20-21 November, 2004. We'll finish by about > 3:00 MST on Sunday, so you could catch a flight as early as about > 5:30. Also, Mark Wilkinson will be giving a talk on MOBY Services at > NCGR on November 19 (some time in the afternoon, I believe); everyone > is welcome to attend, so consider coming a day early. I'll soon put > together a page of information with an agenda, links to accommodations > covering a range of prices, and other details. There will be a small > registration fee to cover breakfast and lunch for Saturday and Sunday. > > In the interim, please post suggestions for topics to cover at the > meeting. > > ---- > > Gary Schiltz > Principal Software Engineer > National Center for Genome Resources > Santa Fe, New Mexico > > gss@ncgr.org > 505-995-4414 (Work) > 505-670-5983 (Cell) > From gss at ncgr.org Thu Oct 7 16:18:18 2004 From: gss at ncgr.org (Gary Schiltz) Date: Thu Oct 7 16:21:20 2004 Subject: [MOBY-dev] MOBY Meeting Registration Message-ID: <4165A48A.5010804@ncgr.org> The online registration form for the MOBY meeting is now available at www.semanticmoby.org/meeting/registration-form.html. Even if you have already sent an email letting me know that you're planning to attend, I'd appreciate it if you would fill out the form anyway; it will help me keep track a bit better. // Gary From gss at ncgr.org Mon Oct 11 15:55:56 2004 From: gss at ncgr.org (Gary Schiltz) Date: Mon Oct 11 15:59:00 2004 Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: References: Message-ID: <416AE54C.9030805@ncgr.org> Martin, I hope it's alright for me to post this to the lists, as it should be of interest to everyone thinking of coming to the MOBY meeting. I found a $99 per night rate through Hotel Santa Fe's web site. Go to http://www.hotelsantafe.com/home.html, click on "Rates & Reservations" on the left side of the window, and in the pop-up window that opens, enter 2004-November 19 for Arrival, 2004-November 21 for Departure, click "Check Availability", and scroll down to the "Hot Internet Rates" section - there should be a "Hotel Santa Fe-Traditional King" rate of $99 per night. That is very good for a nice hotel in Santa Fe. There are, however, cheaper alternatives; I found hotels for under $50 per night on Travelocity. If anyone needs less expensive places to stay, we can arrange for transportation (the cheaper places are a bit further away). Just use MapQuest to check the distance from a given hotel to the Santa Fe Hotel. Looking forward to seeing everyone in November! // Gary Martin Senger wrote: > Gary, > You recommended SanteFe hotel. I thought that you had some sort of > arrangement with them - such as a spacial and reasonable price for BioMoby > meeting participants. Because without such arrangement it does not seem > (at least for us from academia) to be a choice. The cheapest hotel room > they offered us today was for 263 US dollars plus tax ! > Is there a second choice that you can recommend perhaps? As you said, > to stay in the same hotel may help us to socialize and so on. > > Thanks for your advise and with regards, > Martin > From senger at ebi.ac.uk Mon Oct 11 16:11:31 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon Oct 11 16:11:48 2004 Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: <416AE54C.9030805@ncgr.org> Message-ID: > I hope it's alright for me to post this to the lists > Of course. > $99 per night rate through Hotel Santa Fe's web site > Indeed, it is true. The price 263, I mentioned in my email, was the one given to our admin when she asked by phone. Inded a difference. Thanks, and see you there. Martin -- Martin Senger EMBL Outstation - Hinxton Senger@EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From yanwong at ebgm.jussieu.fr Thu Oct 14 09:11:13 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: Thu Oct 14 09:10:27 2004 Subject: [MOBY-dev] How does work the RDF system?! Message-ID: <1097759472.14034.24.camel@prot1.rpbs.jussieu.fr> I got some questions about the new registration process: I'd made a mistake when registering a service (the URL was wrong). I modified the RDF file in order to correct the mistake. The day after, I tried the service again but it didn't take into account the modifications in the RDF... (the URL was still wrong!) Did I miss something? I wanted to deregister a service. I deleted the rdf declaration from the file but several days after, the service was still here! (after a findService query). With the RDF file, does it mean that we can add new services without having to call the register method? After all, we can add the declaration of the service in the file and the service would be automatically registered into the services directory. What should I put in the SignatureURL field? the url of rdf file itself or just the path to it: http:///RDF or http:///RDF/file_containing_signatures.rdf ? From yanwong at ebgm.jussieu.fr Thu Oct 14 09:23:09 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: Thu Oct 14 09:22:10 2004 Subject: [MOBY-dev] the bioMoby python API Message-ID: <1097760189.14174.7.camel@prot1.rpbs.jussieu.fr> I am modifying the API in order to deal with the RDF files and solving some bugs: -Bug in the Dispatcher corrected (troubles with handling multiple queries simultaneously) added signatureURL and RDF for Registration process (in Service, ServiceList) -debug option when executing a service (trace of HTTP calls) in Central and Service Class (allows you to have the dump of all your SOAP transaction with the webservice) if you want to trace a SOAP call while executing a service just put: service.execute(mobyqueries, debug=True) and you'll have the complete trace of the execution when you deal with the Moby Central Server: client=Client(debug=True) to trace execution of your moby Query As Errare Humanum Est, feel free to notify me errors I missed :) I expect to release a corrected version in the end of next week. From senger at ebi.ac.uk Mon Oct 18 11:53:38 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon Oct 18 11:52:39 2004 Subject: [MOBY-dev] moby grapgs updated In-Reply-To: <4165A48A.5010804@ncgr.org> Message-ID: Hi all, I have updated moby graphs (and also few other things in jMoby - see docs/ChangeLog if you wish). The main change is that now there is a possibility to draw a path between two selected data types, and that the result can be expressed as a workflow definition that can be consequently loaded to Taverna and hopefully run. I expect (at least) two things to happen soon: * You will/may find that the caching in graph servlet is not yet perfect. I need some time to make it more robust, but I will be travelling now - so if you find problems, just try the same again, and blame me... * Having now generated workflow definitions, more people will try it in Taverna. This is perfect, it's an ultimate goal. But I again need more time to improve moby stuf in Taverna, again make it more robust. And I will - so again try to send your comments, suggestion and bug reports, and blame me... Also I hope that I fixed few things that Frank was reporting: > 1. MOBY allows multiple whitespace-separated words in the object > ontology, but when they are graphed, they show up as separate nodes. > This is probably just a matter of enclosing the node names in double > quotes for GraphViz. > 2. The first time I try to draw the object ontology, after reloading the > registry, I get an error from Tomcat like this: Finally, here is the URL of the new moby graphs (actually it is the same as before, you can still get there from the main BioMoby page): http://www.ebi.ac.uk/collab/mygrid/service2/jmoby/graphs With regards, Martin -- Martin Senger EMBL Outstation - Hinxton Senger@EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From rebecca.ernst at gsf.de Thu Oct 21 03:16:00 2004 From: rebecca.ernst at gsf.de (Rebecca Ernst) Date: Thu Oct 21 03:14:51 2004 Subject: [MOBY-dev] moby grapgs updated In-Reply-To: References: Message-ID: <41776230.1030108@gsf.de> Hi Martin! that's another nice feature! the problem I have at the moment is that I can't squeeze any graph out of the tool- I always get Apache errors. (I tried all graphs and none worked. the new one is timing out) cheers, rebecca Martin Senger wrote: >Hi all, > I have updated moby graphs (and also few other things in jMoby - see >docs/ChangeLog if you wish). > The main change is that now there is a possibility to draw a path >between two selected data types, and that the result can be expressed as a >workflow definition that can be consequently loaded to Taverna and >hopefully run. > I expect (at least) two things to happen soon: > > * You will/may find that the caching in graph servlet is not yet >perfect. I need some time to make it more robust, but I will be travelling >now - so if you find problems, just try the same again, and blame me... > > * Having now generated workflow definitions, more people will try it in >Taverna. This is perfect, it's an ultimate goal. But I again need more >time to improve moby stuf in Taverna, again make it more robust. And I >will - so again try to send your comments, suggestion and bug reports, and >blame me... > > Also I hope that I fixed few things that Frank was reporting: > > >>1. MOBY allows multiple whitespace-separated words in the object >>ontology, but when they are graphed, they show up as separate nodes. >>This is probably just a matter of enclosing the node names in double >>quotes for GraphViz. >> >> > > > >>2. The first time I try to draw the object ontology, after reloading the >>registry, I get an error from Tomcat like this: >> >> > > Finally, here is the URL of the new moby graphs (actually it is the >same as before, you can still get there from the main BioMoby page): > >http://www.ebi.ac.uk/collab/mygrid/service2/jmoby/graphs > > With regards, > Martin > > > -- Rebecca Ernst MIPS, Inst. for Bioinformatics GSF Research Center for Environment and Health Ingolstaedter Landstr. 1 85764 Neuherberg fon: +49 89 3187 3583 email: Rebecca.Ernst@gsf.de From senger at ebi.ac.uk Thu Oct 21 08:09:27 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Thu Oct 21 08:08:22 2004 Subject: [MOBY-dev] moby grapgs updated In-Reply-To: <41776230.1030108@gsf.de> Message-ID: > that's another nice feature! > the problem I have at the moment is that I can't squeeze any graph out > of the tool- I always get Apache errors. > Sorry, I am aware of the problem, but I am not sure if I can fix it from the Internet coffee in Prague where I am sitting now. Strange kezboard, these Czechs they have here :-) It will be fixed latest on Mondaz and I will let you u know. Again, I am sorry. Martin -- Martin Senger EMBL Outstation - Hinxton Senger@EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From yanwong at ebgm.jussieu.fr Fri Oct 22 09:14:07 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: Fri Oct 22 09:12:56 2004 Subject: [MOBY-dev] bioMoby Python / RDF / ViPEr and bioMoby webservices... Message-ID: <1098450847.15817.32.camel@prot1.rpbs.jussieu.fr> I've put the last version (0.6) of the bioMoby Python API. You can retrieve it on http://bioserv.rpbs.jussieu.fr/software/BioMoby-Python.tar.bz2 here are the changes: Some changes in the documentation -Bug in the MobyObject corrected -Bug in the Dispatcher corrected (troubles with handling multiple queries simultaneously) -Added support for signatureURL and RDF in the registration process -Debug option available when executing a service (trace of HTTP calls) in Class Central and Class Service -Added tutorials on how to build a webservice using openPBS and make asynchronous queries. I am also working on a script that will integrate bioMoby webservices in ViPEr. At this time, I've made some prototypes of interface with ViPEr I wrote some interfaces for some bioMoby webservices. It is rather straightfoward (I may be able to write an automatic ViPEr component generator). Here some screenshots: http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot1.jpg this one uses services from RPBS http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot2.jpg I give him a PDB ID and it get a FASTA sequence with MOBYSHoundGetGenBankFasta and make a BlastFastaVsArabiProteincoding and a query with Automat. http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot5.jpg Results are printed on the stdout as seen on the terminal where I launcher ViPer http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot3.jpg This one uses a function that generates a AA FASTA sequence and a 3D sequence The two sequences are sent to the BasicBuilder webservice. Calculation of ASA with ASA webservice and secondary structure with Stride (RPBS webservices) http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot4.jpg The result of the job. I expect to have good results in next two weeks... A Propos RDF, can someone finally tell me how it works? Does modifications(deletion, addition, modification) in the RDF files modify entries in the bioMoby directory? From senger at ebi.ac.uk Mon Oct 25 08:06:12 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon Oct 25 08:05:02 2004 Subject: [MOBY-dev] moby grapgs updated In-Reply-To: Message-ID: As promised I am trying to find what is the problem with the Moby graphs. At the moment I do not know yet the solution (it is the famous situation that everything works fine on your laptop but stops to work when deployed on the publicly available server) but I will find it. Meanwhile, you can (always) create graphs - if you need them from the command line (assuming that you have Java installed on your computer, and you have a 'dot' program installed). This is a short HOWTO (which I will put later on the jMoby pages): 1) cd /Java 3) cvs update -dP 3) ./build.sh Now you should have everything prepared on your machine. For the graphs try the following: 4) build/run/run-graphs-client -help or (without reading help): 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot 5) dot -Tpng result.dot > result.png You will almost certainly try several times - therefore it's worth to cache results from the Moby Central locally. Do it by adding -cachedir parameter (the first time it will fill this directory, next time it will reuse it): 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot -cachedir /tmp/mycache Finally if you want to see also scufl definitions try this: 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot -cachedir /tmp/mycache \ -scufl I will let you know when I solve the problem with my server - so you will be able to get graphs in a more convenient way (but the way described above should work always, disregarding our server down etc. so it is perhaps worth to know about it I guess). Regards, Martin -- Martin Senger EMBL Outstation - Hinxton Senger@EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From edward.kawas at gmail.com Sun Oct 3 00:09:33 2004 From: edward.kawas at gmail.com (Eddie Kawas) Date: Sat, 2 Oct 2004 21:09:33 -0700 Subject: [MOBY-dev] Tools for creating service types, service instances, and objects In-Reply-To: <4155C12D.6020305@illuminae.com> References: <4155C12D.6020305@illuminae.com> Message-ID: Hi, I have created some new tools to aid in the construction of service types, service instances, and objects. This is the first release of the tools and I was hoping to see what people think. Anyways, the url for: *objects http://mobycentral.cbr.nrc.ca/applets/object.html *service instances http://mobycentral.cbr.nrc.ca/applets/service.html *service types http://mobycentral.cbr.nrc.ca/applets/servicetype.html The tools are java applets and are 'live'. The initial download is quite large because of some libraries that are needed. Eddie From gss at ncgr.org Wed Oct 6 16:14:26 2004 From: gss at ncgr.org (Gary Schiltz) Date: Wed, 06 Oct 2004 14:14:26 -0600 Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: <415440E3.2070000@ncgr.org> References: <415440E3.2070000@ncgr.org> Message-ID: <41645222.1060907@ncgr.org> I've put information about the MOBY meeting on the Semantic MOBY web site - the URL is www.semanticmoby.org/meeting. It is currently missing a registration form, which I'll add by the end of the day. // Gary Gary Schiltz wrote: > Hello MOBY'ers! > > The next MOBY meeting will be held at NCGR (www.ncgr.org) in Santa Fe, > New Mexico, the weekend of 20-21 November, 2004. We'll finish by about > 3:00 MST on Sunday, so you could catch a flight as early as about > 5:30. Also, Mark Wilkinson will be giving a talk on MOBY Services at > NCGR on November 19 (some time in the afternoon, I believe); everyone > is welcome to attend, so consider coming a day early. I'll soon put > together a page of information with an agenda, links to accommodations > covering a range of prices, and other details. There will be a small > registration fee to cover breakfast and lunch for Saturday and Sunday. > > In the interim, please post suggestions for topics to cover at the > meeting. > > ---- > > Gary Schiltz > Principal Software Engineer > National Center for Genome Resources > Santa Fe, New Mexico > > gss at ncgr.org > 505-995-4414 (Work) > 505-670-5983 (Cell) > From gss at ncgr.org Thu Oct 7 16:18:18 2004 From: gss at ncgr.org (Gary Schiltz) Date: Thu, 07 Oct 2004 14:18:18 -0600 Subject: [MOBY-dev] MOBY Meeting Registration Message-ID: <4165A48A.5010804@ncgr.org> The online registration form for the MOBY meeting is now available at www.semanticmoby.org/meeting/registration-form.html. Even if you have already sent an email letting me know that you're planning to attend, I'd appreciate it if you would fill out the form anyway; it will help me keep track a bit better. // Gary From gss at ncgr.org Mon Oct 11 15:55:56 2004 From: gss at ncgr.org (Gary Schiltz) Date: Mon, 11 Oct 2004 13:55:56 -0600 Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: References: Message-ID: <416AE54C.9030805@ncgr.org> Martin, I hope it's alright for me to post this to the lists, as it should be of interest to everyone thinking of coming to the MOBY meeting. I found a $99 per night rate through Hotel Santa Fe's web site. Go to http://www.hotelsantafe.com/home.html, click on "Rates & Reservations" on the left side of the window, and in the pop-up window that opens, enter 2004-November 19 for Arrival, 2004-November 21 for Departure, click "Check Availability", and scroll down to the "Hot Internet Rates" section - there should be a "Hotel Santa Fe-Traditional King" rate of $99 per night. That is very good for a nice hotel in Santa Fe. There are, however, cheaper alternatives; I found hotels for under $50 per night on Travelocity. If anyone needs less expensive places to stay, we can arrange for transportation (the cheaper places are a bit further away). Just use MapQuest to check the distance from a given hotel to the Santa Fe Hotel. Looking forward to seeing everyone in November! // Gary Martin Senger wrote: > Gary, > You recommended SanteFe hotel. I thought that you had some sort of > arrangement with them - such as a spacial and reasonable price for BioMoby > meeting participants. Because without such arrangement it does not seem > (at least for us from academia) to be a choice. The cheapest hotel room > they offered us today was for 263 US dollars plus tax ! > Is there a second choice that you can recommend perhaps? As you said, > to stay in the same hotel may help us to socialize and so on. > > Thanks for your advise and with regards, > Martin > From senger at ebi.ac.uk Mon Oct 11 16:11:31 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon, 11 Oct 2004 21:11:31 +0100 (BST) Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: <416AE54C.9030805@ncgr.org> Message-ID: > I hope it's alright for me to post this to the lists > Of course. > $99 per night rate through Hotel Santa Fe's web site > Indeed, it is true. The price 263, I mentioned in my email, was the one given to our admin when she asked by phone. Inded a difference. Thanks, and see you there. Martin -- Martin Senger EMBL Outstation - Hinxton Senger at EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From yanwong at ebgm.jussieu.fr Thu Oct 14 09:11:13 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: 14 Oct 2004 15:11:13 +0200 Subject: [MOBY-dev] How does work the RDF system?! Message-ID: <1097759472.14034.24.camel@prot1.rpbs.jussieu.fr> I got some questions about the new registration process: I'd made a mistake when registering a service (the URL was wrong). I modified the RDF file in order to correct the mistake. The day after, I tried the service again but it didn't take into account the modifications in the RDF... (the URL was still wrong!) Did I miss something? I wanted to deregister a service. I deleted the rdf declaration from the file but several days after, the service was still here! (after a findService query). With the RDF file, does it mean that we can add new services without having to call the register method? After all, we can add the declaration of the service in the file and the service would be automatically registered into the services directory. What should I put in the SignatureURL field? the url of rdf file itself or just the path to it: http:///RDF or http:///RDF/file_containing_signatures.rdf ? From yanwong at ebgm.jussieu.fr Thu Oct 14 09:23:09 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: 14 Oct 2004 15:23:09 +0200 Subject: [MOBY-dev] the bioMoby python API Message-ID: <1097760189.14174.7.camel@prot1.rpbs.jussieu.fr> I am modifying the API in order to deal with the RDF files and solving some bugs: -Bug in the Dispatcher corrected (troubles with handling multiple queries simultaneously) added signatureURL and RDF for Registration process (in Service, ServiceList) -debug option when executing a service (trace of HTTP calls) in Central and Service Class (allows you to have the dump of all your SOAP transaction with the webservice) if you want to trace a SOAP call while executing a service just put: service.execute(mobyqueries, debug=True) and you'll have the complete trace of the execution when you deal with the Moby Central Server: client=Client(debug=True) to trace execution of your moby Query As Errare Humanum Est, feel free to notify me errors I missed :) I expect to release a corrected version in the end of next week. From senger at ebi.ac.uk Mon Oct 18 11:53:38 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon, 18 Oct 2004 16:53:38 +0100 (BST) Subject: [MOBY-dev] moby grapgs updated In-Reply-To: <4165A48A.5010804@ncgr.org> Message-ID: Hi all, I have updated moby graphs (and also few other things in jMoby - see docs/ChangeLog if you wish). The main change is that now there is a possibility to draw a path between two selected data types, and that the result can be expressed as a workflow definition that can be consequently loaded to Taverna and hopefully run. I expect (at least) two things to happen soon: * You will/may find that the caching in graph servlet is not yet perfect. I need some time to make it more robust, but I will be travelling now - so if you find problems, just try the same again, and blame me... * Having now generated workflow definitions, more people will try it in Taverna. This is perfect, it's an ultimate goal. But I again need more time to improve moby stuf in Taverna, again make it more robust. And I will - so again try to send your comments, suggestion and bug reports, and blame me... Also I hope that I fixed few things that Frank was reporting: > 1. MOBY allows multiple whitespace-separated words in the object > ontology, but when they are graphed, they show up as separate nodes. > This is probably just a matter of enclosing the node names in double > quotes for GraphViz. > 2. The first time I try to draw the object ontology, after reloading the > registry, I get an error from Tomcat like this: Finally, here is the URL of the new moby graphs (actually it is the same as before, you can still get there from the main BioMoby page): http://www.ebi.ac.uk/collab/mygrid/service2/jmoby/graphs With regards, Martin -- Martin Senger EMBL Outstation - Hinxton Senger at EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From rebecca.ernst at gsf.de Thu Oct 21 03:16:00 2004 From: rebecca.ernst at gsf.de (Rebecca Ernst) Date: Thu, 21 Oct 2004 09:16:00 +0200 Subject: [MOBY-dev] moby grapgs updated In-Reply-To: References: Message-ID: <41776230.1030108@gsf.de> Hi Martin! that's another nice feature! the problem I have at the moment is that I can't squeeze any graph out of the tool- I always get Apache errors. (I tried all graphs and none worked. the new one is timing out) cheers, rebecca Martin Senger wrote: >Hi all, > I have updated moby graphs (and also few other things in jMoby - see >docs/ChangeLog if you wish). > The main change is that now there is a possibility to draw a path >between two selected data types, and that the result can be expressed as a >workflow definition that can be consequently loaded to Taverna and >hopefully run. > I expect (at least) two things to happen soon: > > * You will/may find that the caching in graph servlet is not yet >perfect. I need some time to make it more robust, but I will be travelling >now - so if you find problems, just try the same again, and blame me... > > * Having now generated workflow definitions, more people will try it in >Taverna. This is perfect, it's an ultimate goal. But I again need more >time to improve moby stuf in Taverna, again make it more robust. And I >will - so again try to send your comments, suggestion and bug reports, and >blame me... > > Also I hope that I fixed few things that Frank was reporting: > > >>1. MOBY allows multiple whitespace-separated words in the object >>ontology, but when they are graphed, they show up as separate nodes. >>This is probably just a matter of enclosing the node names in double >>quotes for GraphViz. >> >> > > > >>2. The first time I try to draw the object ontology, after reloading the >>registry, I get an error from Tomcat like this: >> >> > > Finally, here is the URL of the new moby graphs (actually it is the >same as before, you can still get there from the main BioMoby page): > >http://www.ebi.ac.uk/collab/mygrid/service2/jmoby/graphs > > With regards, > Martin > > > -- Rebecca Ernst MIPS, Inst. for Bioinformatics GSF Research Center for Environment and Health Ingolstaedter Landstr. 1 85764 Neuherberg fon: +49 89 3187 3583 email: Rebecca.Ernst at gsf.de From senger at ebi.ac.uk Thu Oct 21 08:09:27 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Thu, 21 Oct 2004 13:09:27 +0100 (BST) Subject: [MOBY-dev] moby grapgs updated In-Reply-To: <41776230.1030108@gsf.de> Message-ID: > that's another nice feature! > the problem I have at the moment is that I can't squeeze any graph out > of the tool- I always get Apache errors. > Sorry, I am aware of the problem, but I am not sure if I can fix it from the Internet coffee in Prague where I am sitting now. Strange kezboard, these Czechs they have here :-) It will be fixed latest on Mondaz and I will let you u know. Again, I am sorry. Martin -- Martin Senger EMBL Outstation - Hinxton Senger at EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From yanwong at ebgm.jussieu.fr Fri Oct 22 09:14:07 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: 22 Oct 2004 15:14:07 +0200 Subject: [MOBY-dev] bioMoby Python / RDF / ViPEr and bioMoby webservices... Message-ID: <1098450847.15817.32.camel@prot1.rpbs.jussieu.fr> I've put the last version (0.6) of the bioMoby Python API. You can retrieve it on http://bioserv.rpbs.jussieu.fr/software/BioMoby-Python.tar.bz2 here are the changes: Some changes in the documentation -Bug in the MobyObject corrected -Bug in the Dispatcher corrected (troubles with handling multiple queries simultaneously) -Added support for signatureURL and RDF in the registration process -Debug option available when executing a service (trace of HTTP calls) in Class Central and Class Service -Added tutorials on how to build a webservice using openPBS and make asynchronous queries. I am also working on a script that will integrate bioMoby webservices in ViPEr. At this time, I've made some prototypes of interface with ViPEr I wrote some interfaces for some bioMoby webservices. It is rather straightfoward (I may be able to write an automatic ViPEr component generator). Here some screenshots: http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot1.jpg this one uses services from RPBS http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot2.jpg I give him a PDB ID and it get a FASTA sequence with MOBYSHoundGetGenBankFasta and make a BlastFastaVsArabiProteincoding and a query with Automat. http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot5.jpg Results are printed on the stdout as seen on the terminal where I launcher ViPer http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot3.jpg This one uses a function that generates a AA FASTA sequence and a 3D sequence The two sequences are sent to the BasicBuilder webservice. Calculation of ASA with ASA webservice and secondary structure with Stride (RPBS webservices) http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot4.jpg The result of the job. I expect to have good results in next two weeks... A Propos RDF, can someone finally tell me how it works? Does modifications(deletion, addition, modification) in the RDF files modify entries in the bioMoby directory? From senger at ebi.ac.uk Mon Oct 25 08:06:12 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon, 25 Oct 2004 13:06:12 +0100 (BST) Subject: [MOBY-dev] moby grapgs updated In-Reply-To: Message-ID: As promised I am trying to find what is the problem with the Moby graphs. At the moment I do not know yet the solution (it is the famous situation that everything works fine on your laptop but stops to work when deployed on the publicly available server) but I will find it. Meanwhile, you can (always) create graphs - if you need them from the command line (assuming that you have Java installed on your computer, and you have a 'dot' program installed). This is a short HOWTO (which I will put later on the jMoby pages): 1) cd /Java 3) cvs update -dP 3) ./build.sh Now you should have everything prepared on your machine. For the graphs try the following: 4) build/run/run-graphs-client -help or (without reading help): 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot 5) dot -Tpng result.dot > result.png You will almost certainly try several times - therefore it's worth to cache results from the Moby Central locally. Do it by adding -cachedir parameter (the first time it will fill this directory, next time it will reuse it): 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot -cachedir /tmp/mycache Finally if you want to see also scufl definitions try this: 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot -cachedir /tmp/mycache \ -scufl I will let you know when I solve the problem with my server - so you will be able to get graphs in a more convenient way (but the way described above should work always, disregarding our server down etc. so it is perhaps worth to know about it I guess). Regards, Martin -- Martin Senger EMBL Outstation - Hinxton Senger at EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From edward.kawas at gmail.com Sun Oct 3 04:09:33 2004 From: edward.kawas at gmail.com (Eddie Kawas) Date: Sat, 2 Oct 2004 21:09:33 -0700 Subject: [MOBY-dev] Tools for creating service types, service instances, and objects In-Reply-To: <4155C12D.6020305@illuminae.com> References: <4155C12D.6020305@illuminae.com> Message-ID: Hi, I have created some new tools to aid in the construction of service types, service instances, and objects. This is the first release of the tools and I was hoping to see what people think. Anyways, the url for: *objects http://mobycentral.cbr.nrc.ca/applets/object.html *service instances http://mobycentral.cbr.nrc.ca/applets/service.html *service types http://mobycentral.cbr.nrc.ca/applets/servicetype.html The tools are java applets and are 'live'. The initial download is quite large because of some libraries that are needed. Eddie From gss at ncgr.org Wed Oct 6 20:14:26 2004 From: gss at ncgr.org (Gary Schiltz) Date: Wed, 06 Oct 2004 14:14:26 -0600 Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: <415440E3.2070000@ncgr.org> References: <415440E3.2070000@ncgr.org> Message-ID: <41645222.1060907@ncgr.org> I've put information about the MOBY meeting on the Semantic MOBY web site - the URL is www.semanticmoby.org/meeting. It is currently missing a registration form, which I'll add by the end of the day. // Gary Gary Schiltz wrote: > Hello MOBY'ers! > > The next MOBY meeting will be held at NCGR (www.ncgr.org) in Santa Fe, > New Mexico, the weekend of 20-21 November, 2004. We'll finish by about > 3:00 MST on Sunday, so you could catch a flight as early as about > 5:30. Also, Mark Wilkinson will be giving a talk on MOBY Services at > NCGR on November 19 (some time in the afternoon, I believe); everyone > is welcome to attend, so consider coming a day early. I'll soon put > together a page of information with an agenda, links to accommodations > covering a range of prices, and other details. There will be a small > registration fee to cover breakfast and lunch for Saturday and Sunday. > > In the interim, please post suggestions for topics to cover at the > meeting. > > ---- > > Gary Schiltz > Principal Software Engineer > National Center for Genome Resources > Santa Fe, New Mexico > > gss at ncgr.org > 505-995-4414 (Work) > 505-670-5983 (Cell) > From gss at ncgr.org Thu Oct 7 20:18:18 2004 From: gss at ncgr.org (Gary Schiltz) Date: Thu, 07 Oct 2004 14:18:18 -0600 Subject: [MOBY-dev] MOBY Meeting Registration Message-ID: <4165A48A.5010804@ncgr.org> The online registration form for the MOBY meeting is now available at www.semanticmoby.org/meeting/registration-form.html. Even if you have already sent an email letting me know that you're planning to attend, I'd appreciate it if you would fill out the form anyway; it will help me keep track a bit better. // Gary From gss at ncgr.org Mon Oct 11 19:55:56 2004 From: gss at ncgr.org (Gary Schiltz) Date: Mon, 11 Oct 2004 13:55:56 -0600 Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: References: Message-ID: <416AE54C.9030805@ncgr.org> Martin, I hope it's alright for me to post this to the lists, as it should be of interest to everyone thinking of coming to the MOBY meeting. I found a $99 per night rate through Hotel Santa Fe's web site. Go to http://www.hotelsantafe.com/home.html, click on "Rates & Reservations" on the left side of the window, and in the pop-up window that opens, enter 2004-November 19 for Arrival, 2004-November 21 for Departure, click "Check Availability", and scroll down to the "Hot Internet Rates" section - there should be a "Hotel Santa Fe-Traditional King" rate of $99 per night. That is very good for a nice hotel in Santa Fe. There are, however, cheaper alternatives; I found hotels for under $50 per night on Travelocity. If anyone needs less expensive places to stay, we can arrange for transportation (the cheaper places are a bit further away). Just use MapQuest to check the distance from a given hotel to the Santa Fe Hotel. Looking forward to seeing everyone in November! // Gary Martin Senger wrote: > Gary, > You recommended SanteFe hotel. I thought that you had some sort of > arrangement with them - such as a spacial and reasonable price for BioMoby > meeting participants. Because without such arrangement it does not seem > (at least for us from academia) to be a choice. The cheapest hotel room > they offered us today was for 263 US dollars plus tax ! > Is there a second choice that you can recommend perhaps? As you said, > to stay in the same hotel may help us to socialize and so on. > > Thanks for your advise and with regards, > Martin > From senger at ebi.ac.uk Mon Oct 11 20:11:31 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon, 11 Oct 2004 21:11:31 +0100 (BST) Subject: [MOBY-dev] Re: MOBY Meeting: 20-21 November in Santa Fe In-Reply-To: <416AE54C.9030805@ncgr.org> Message-ID: > I hope it's alright for me to post this to the lists > Of course. > $99 per night rate through Hotel Santa Fe's web site > Indeed, it is true. The price 263, I mentioned in my email, was the one given to our admin when she asked by phone. Inded a difference. Thanks, and see you there. Martin -- Martin Senger EMBL Outstation - Hinxton Senger at EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From yanwong at ebgm.jussieu.fr Thu Oct 14 13:11:13 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: 14 Oct 2004 15:11:13 +0200 Subject: [MOBY-dev] How does work the RDF system?! Message-ID: <1097759472.14034.24.camel@prot1.rpbs.jussieu.fr> I got some questions about the new registration process: I'd made a mistake when registering a service (the URL was wrong). I modified the RDF file in order to correct the mistake. The day after, I tried the service again but it didn't take into account the modifications in the RDF... (the URL was still wrong!) Did I miss something? I wanted to deregister a service. I deleted the rdf declaration from the file but several days after, the service was still here! (after a findService query). With the RDF file, does it mean that we can add new services without having to call the register method? After all, we can add the declaration of the service in the file and the service would be automatically registered into the services directory. What should I put in the SignatureURL field? the url of rdf file itself or just the path to it: http:///RDF or http:///RDF/file_containing_signatures.rdf ? From yanwong at ebgm.jussieu.fr Thu Oct 14 13:23:09 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: 14 Oct 2004 15:23:09 +0200 Subject: [MOBY-dev] the bioMoby python API Message-ID: <1097760189.14174.7.camel@prot1.rpbs.jussieu.fr> I am modifying the API in order to deal with the RDF files and solving some bugs: -Bug in the Dispatcher corrected (troubles with handling multiple queries simultaneously) added signatureURL and RDF for Registration process (in Service, ServiceList) -debug option when executing a service (trace of HTTP calls) in Central and Service Class (allows you to have the dump of all your SOAP transaction with the webservice) if you want to trace a SOAP call while executing a service just put: service.execute(mobyqueries, debug=True) and you'll have the complete trace of the execution when you deal with the Moby Central Server: client=Client(debug=True) to trace execution of your moby Query As Errare Humanum Est, feel free to notify me errors I missed :) I expect to release a corrected version in the end of next week. From senger at ebi.ac.uk Mon Oct 18 15:53:38 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon, 18 Oct 2004 16:53:38 +0100 (BST) Subject: [MOBY-dev] moby grapgs updated In-Reply-To: <4165A48A.5010804@ncgr.org> Message-ID: Hi all, I have updated moby graphs (and also few other things in jMoby - see docs/ChangeLog if you wish). The main change is that now there is a possibility to draw a path between two selected data types, and that the result can be expressed as a workflow definition that can be consequently loaded to Taverna and hopefully run. I expect (at least) two things to happen soon: * You will/may find that the caching in graph servlet is not yet perfect. I need some time to make it more robust, but I will be travelling now - so if you find problems, just try the same again, and blame me... * Having now generated workflow definitions, more people will try it in Taverna. This is perfect, it's an ultimate goal. But I again need more time to improve moby stuf in Taverna, again make it more robust. And I will - so again try to send your comments, suggestion and bug reports, and blame me... Also I hope that I fixed few things that Frank was reporting: > 1. MOBY allows multiple whitespace-separated words in the object > ontology, but when they are graphed, they show up as separate nodes. > This is probably just a matter of enclosing the node names in double > quotes for GraphViz. > 2. The first time I try to draw the object ontology, after reloading the > registry, I get an error from Tomcat like this: Finally, here is the URL of the new moby graphs (actually it is the same as before, you can still get there from the main BioMoby page): http://www.ebi.ac.uk/collab/mygrid/service2/jmoby/graphs With regards, Martin -- Martin Senger EMBL Outstation - Hinxton Senger at EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From rebecca.ernst at gsf.de Thu Oct 21 07:16:00 2004 From: rebecca.ernst at gsf.de (Rebecca Ernst) Date: Thu, 21 Oct 2004 09:16:00 +0200 Subject: [MOBY-dev] moby grapgs updated In-Reply-To: References: Message-ID: <41776230.1030108@gsf.de> Hi Martin! that's another nice feature! the problem I have at the moment is that I can't squeeze any graph out of the tool- I always get Apache errors. (I tried all graphs and none worked. the new one is timing out) cheers, rebecca Martin Senger wrote: >Hi all, > I have updated moby graphs (and also few other things in jMoby - see >docs/ChangeLog if you wish). > The main change is that now there is a possibility to draw a path >between two selected data types, and that the result can be expressed as a >workflow definition that can be consequently loaded to Taverna and >hopefully run. > I expect (at least) two things to happen soon: > > * You will/may find that the caching in graph servlet is not yet >perfect. I need some time to make it more robust, but I will be travelling >now - so if you find problems, just try the same again, and blame me... > > * Having now generated workflow definitions, more people will try it in >Taverna. This is perfect, it's an ultimate goal. But I again need more >time to improve moby stuf in Taverna, again make it more robust. And I >will - so again try to send your comments, suggestion and bug reports, and >blame me... > > Also I hope that I fixed few things that Frank was reporting: > > >>1. MOBY allows multiple whitespace-separated words in the object >>ontology, but when they are graphed, they show up as separate nodes. >>This is probably just a matter of enclosing the node names in double >>quotes for GraphViz. >> >> > > > >>2. The first time I try to draw the object ontology, after reloading the >>registry, I get an error from Tomcat like this: >> >> > > Finally, here is the URL of the new moby graphs (actually it is the >same as before, you can still get there from the main BioMoby page): > >http://www.ebi.ac.uk/collab/mygrid/service2/jmoby/graphs > > With regards, > Martin > > > -- Rebecca Ernst MIPS, Inst. for Bioinformatics GSF Research Center for Environment and Health Ingolstaedter Landstr. 1 85764 Neuherberg fon: +49 89 3187 3583 email: Rebecca.Ernst at gsf.de From senger at ebi.ac.uk Thu Oct 21 12:09:27 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Thu, 21 Oct 2004 13:09:27 +0100 (BST) Subject: [MOBY-dev] moby grapgs updated In-Reply-To: <41776230.1030108@gsf.de> Message-ID: > that's another nice feature! > the problem I have at the moment is that I can't squeeze any graph out > of the tool- I always get Apache errors. > Sorry, I am aware of the problem, but I am not sure if I can fix it from the Internet coffee in Prague where I am sitting now. Strange kezboard, these Czechs they have here :-) It will be fixed latest on Mondaz and I will let you u know. Again, I am sorry. Martin -- Martin Senger EMBL Outstation - Hinxton Senger at EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger From yanwong at ebgm.jussieu.fr Fri Oct 22 13:14:07 2004 From: yanwong at ebgm.jussieu.fr (Yan Wong) Date: 22 Oct 2004 15:14:07 +0200 Subject: [MOBY-dev] bioMoby Python / RDF / ViPEr and bioMoby webservices... Message-ID: <1098450847.15817.32.camel@prot1.rpbs.jussieu.fr> I've put the last version (0.6) of the bioMoby Python API. You can retrieve it on http://bioserv.rpbs.jussieu.fr/software/BioMoby-Python.tar.bz2 here are the changes: Some changes in the documentation -Bug in the MobyObject corrected -Bug in the Dispatcher corrected (troubles with handling multiple queries simultaneously) -Added support for signatureURL and RDF in the registration process -Debug option available when executing a service (trace of HTTP calls) in Class Central and Class Service -Added tutorials on how to build a webservice using openPBS and make asynchronous queries. I am also working on a script that will integrate bioMoby webservices in ViPEr. At this time, I've made some prototypes of interface with ViPEr I wrote some interfaces for some bioMoby webservices. It is rather straightfoward (I may be able to write an automatic ViPEr component generator). Here some screenshots: http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot1.jpg this one uses services from RPBS http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot2.jpg I give him a PDB ID and it get a FASTA sequence with MOBYSHoundGetGenBankFasta and make a BlastFastaVsArabiProteincoding and a query with Automat. http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot5.jpg Results are printed on the stdout as seen on the terminal where I launcher ViPer http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot3.jpg This one uses a function that generates a AA FASTA sequence and a 3D sequence The two sequences are sent to the BasicBuilder webservice. Calculation of ASA with ASA webservice and secondary structure with Stride (RPBS webservices) http://bioserv.rpbs.jussieu.fr/RPBS/images/screenshot4.jpg The result of the job. I expect to have good results in next two weeks... A Propos RDF, can someone finally tell me how it works? Does modifications(deletion, addition, modification) in the RDF files modify entries in the bioMoby directory? From senger at ebi.ac.uk Mon Oct 25 12:06:12 2004 From: senger at ebi.ac.uk (Martin Senger) Date: Mon, 25 Oct 2004 13:06:12 +0100 (BST) Subject: [MOBY-dev] moby grapgs updated In-Reply-To: Message-ID: As promised I am trying to find what is the problem with the Moby graphs. At the moment I do not know yet the solution (it is the famous situation that everything works fine on your laptop but stops to work when deployed on the publicly available server) but I will find it. Meanwhile, you can (always) create graphs - if you need them from the command line (assuming that you have Java installed on your computer, and you have a 'dot' program installed). This is a short HOWTO (which I will put later on the jMoby pages): 1) cd /Java 3) cvs update -dP 3) ./build.sh Now you should have everything prepared on your machine. For the graphs try the following: 4) build/run/run-graphs-client -help or (without reading help): 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot 5) dot -Tpng result.dot > result.png You will almost certainly try several times - therefore it's worth to cache results from the Moby Central locally. Do it by adding -cachedir parameter (the first time it will fill this directory, next time it will reuse it): 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot -cachedir /tmp/mycache Finally if you want to see also scufl definitions try this: 4) build/run/run-graphs-client -s -datapath GO/Object AminoAcidSequence \ -f result.dot -cachedir /tmp/mycache \ -scufl I will let you know when I solve the problem with my server - so you will be able to get graphs in a more convenient way (but the way described above should work always, disregarding our server down etc. so it is perhaps worth to know about it I guess). Regards, Martin -- Martin Senger EMBL Outstation - Hinxton Senger at EBI.ac.uk European Bioinformatics Institute Phone: (+44) 1223 494636 Wellcome Trust Genome Campus (Switchboard: 494444) Hinxton Fax : (+44) 1223 494468 Cambridge CB10 1SD United Kingdom http://industry.ebi.ac.uk/~senger