From phillip.lord at newcastle.ac.uk Thu Jul 3 13:10:28 2008 From: phillip.lord at newcastle.ac.uk (Phillip Lord) Date: Thu, 03 Jul 2008 18:10:28 +0100 Subject: [MOBY-l] Bio-Ontologies SIG, ISMB 2008 - Programme available and registration open Message-ID: - Apologies for cross posting - Bio-Ontologies 2008: Knowledge in Biology - Sunday 20th July- a SIG at Intelligent Systems for Molecular Biology (ISMB) 2008 *^**^***^****^****^****^****^****^****^****^****^****^*****^**^***^** **THE PROGRAMME IS NOW AVAILABLE** http://bio-ontologies.org.uk/download/Bio-Ontologies2008.pdf **REGISTRATION IS OPEN** http://www.iscb.org/ismb2008/registration.php **PAPERS FROM LAST YEAR - PUBLISHED** http://www.biomedcentral.com/1471-2105/9?issue=S5 The best papers from this year will also be published in BMC Bioinformatics. See you in Toronto! The Program Chairs: Phillip Lord (1), Susanna-A. Sansone (2), Nigam Shah (3), Matt Cockerill (4) 1. School of Computing Science, University of Newcastle, UK 2. EMBL-EBI The European Bioinformatics Institute, Cambridge, UK 3. Center for Biomedical Informatics, Stanford University, USA 4. BioMedCentral *^**^***^****^****^****^****^****^****^****^****^****^*****^**^***^** The Bio-Ontologies Meeting has existed as a Special Interest Group meeting at ISMB for more than a decade, making it one of the longest running SIG meetings. The Bio-Ontologies SIG meeting has provided a forum for discussion on the latest and most cutting edge research on ontologies. In this decade, the use of ontologies has become mature, moving from niche to mainstream usage within bioinformatics. As result, this year we are broadening the scope of SIG to include formal and informal approaches to organising, presenting and disseminating knowledge in biology. Keynote This year's keynote speaker will be Professor Philip E. Bourne, from the Skaggs School of Pharmacy & Pharmaceutical Sciences, UCSD. Panel Session This year's panel session will have the following members: - Philip E. Bourne, UCSD - Helen Parkinson, European Bioinformatics Institute - Matt Cockerill, BioMed Central - Mark Wilkinson, University of British Columbia Programme Committee We also wish to thank the programme committee for their excellent input and guidance - the programme committee, organised alphabetically is: Mike Bada, University of Colorado Judith Blake, Jackson Laboratory Frank Gibson, Newcastle University Cliff Joslyn, Pacific National Laboratory Wacek Kusnierczyk, Norwegian University of Science and Technology Robin MacEntire, GSK James Malone, EBI Helen Parkinson, EBI Daniel Rubin, Stanford University Alan Ruttenberg, Science Commons Susie M. Stephens, Eli Lilly Robert Stevens, University of Manchester and the conference organisers. -- -- Phillip Lord, Phone: +44 (0) 191 222 7827 Lecturer in Bioinformatics, Email: phillip.lord at newcastle.ac.uk School of Computing Science, http://homepages.cs.ncl.ac.uk/phillip.lord Claremont Tower Room 909, skype: russet_apples Newcastle University, msn: msn at russet.org.uk NE1 7RU From vmgcastellanos at yahoo.com Fri Jul 18 09:52:04 2008 From: vmgcastellanos at yahoo.com (Victor Manuel) Date: Fri, 18 Jul 2008 06:52:04 -0700 (PDT) Subject: [MOBY-l] Services inputs and outputs Message-ID: <437685.19667.qm@web57005.mail.re3.yahoo.com> Is there an ontology describing biomoby inputs and outputs? does biomoby provide any web service discovery mechanism? does biomoby standardizes the description of web services in bionfo? - > _______________________________________________ moby-l mailing list moby-l at lists.open-bio.org http://lists.open-bio.org/mailman/listinfo/moby-l From pieter.neerincx at gmail.com Fri Jul 18 10:30:10 2008 From: pieter.neerincx at gmail.com (Pieter Neerincx) Date: Fri, 18 Jul 2008 16:30:10 +0200 Subject: [MOBY-l] Services inputs and outputs In-Reply-To: <437685.19667.qm@web57005.mail.re3.yahoo.com> References: <437685.19667.qm@web57005.mail.re3.yahoo.com> Message-ID: <6665B46F-B7F3-42A1-BEA7-FB8CFCB30929@gmail.com> Hi Victor, On 18?Jul?2008, at 3:52 PM, Victor Manuel wrote: > Is there an ontology describing biomoby inputs and outputs? Yes, that is what BioMoby was designed for in the first place :) > does biomoby provide any web service discovery mechanism? Yes, that is what the BioMoby ontologies mentioned above were designed for in the first place :) > does biomoby standardizes the description of web services in bionfo? Partially, that is what BioMoby was designed for in the first place :) The inputs and outputs are standardised as well as service types, but there is also a "free text form" description field to describe a service. The latter is just meant to be a human readable string and is not standardised (not even on language, so you might find some non- english stuff in there...) Both the inputs & outputs as well as service types and the services themselves are stored in ontologies. These are available from repositories called BioMoby Central. There is one official public one and several institutes host their own BioMoby Centrals either for development or for services which are not publicly available. Looks like you could use a BioMoby primer; the BioMoby 1.0 paper would be good start: http://bib.oxfordjournals.org/cgi/content/abstract/9/3/220 Cheers, Pi > > > > - >> > > _______________________________________________ > moby-l mailing list > moby-l at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/moby-l > > > > > _______________________________________________ > moby-l mailing list > moby-l at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/moby-l ------------------------------------------------------------- Wageningen University and Research centre (WUR) Laboratory of Bioinformatics Transitorium (building 312) room 1034 Dreijenlaan 3 6703 HA Wageningen The Netherlands phone: +31 (0)317-483 060 mobile: +31 (0)6-143 66 783 e-mail: pieter.neerincx at gmail.com skype: pieter.online ------------------------------------------------------------ From markw at illuminae.com Fri Jul 18 11:01:20 2008 From: markw at illuminae.com (Mark Wilkinson) Date: Fri, 18 Jul 2008 08:01:20 -0700 Subject: [MOBY-l] Services inputs and outputs In-Reply-To: <6665B46F-B7F3-42A1-BEA7-FB8CFCB30929@gmail.com> References: <437685.19667.qm@web57005.mail.re3.yahoo.com> <6665B46F-B7F3-42A1-BEA7-FB8CFCB30929@gmail.com> Message-ID: Thanks Pieter! I'm sitting in the BOSC meeting at ISMB and didn't have the opportunity for a full response :-) M On Fri, 18 Jul 2008 07:30:10 -0700, Pieter Neerincx wrote: > Hi Victor, > > On 18?Jul?2008, at 3:52 PM, Victor Manuel wrote: > >> Is there an ontology describing biomoby inputs and outputs? > > Yes, that is what BioMoby was designed for in the first place :) > >> does biomoby provide any web service discovery mechanism? > > Yes, that is what the BioMoby ontologies mentioned above were designed > for in the first place :) > >> does biomoby standardizes the description of web services in bionfo? > > Partially, that is what BioMoby was designed for in the first place :) > The inputs and outputs are standardised as well as service types, but > there is also a "free text form" description field to describe a > service. The latter is just meant to be a human readable string and is > not standardised (not even on language, so you might find some non- > english stuff in there...) > > Both the inputs & outputs as well as service types and the services > themselves are stored in ontologies. These are available from > repositories called BioMoby Central. There is one official public one > and several institutes host their own BioMoby Centrals either for > development or for services which are not publicly available. > > Looks like you could use a BioMoby primer; the BioMoby 1.0 paper would > be good start: > > http://bib.oxfordjournals.org/cgi/content/abstract/9/3/220 > > Cheers, > > Pi > > >> >> >> >> - >>> >> >> _______________________________________________ >> moby-l mailing list >> moby-l at lists.open-bio.org >> http://lists.open-bio.org/mailman/listinfo/moby-l >> >> >> >> >> _______________________________________________ >> moby-l mailing list >> moby-l at lists.open-bio.org >> http://lists.open-bio.org/mailman/listinfo/moby-l > > ------------------------------------------------------------- > Wageningen University and Research centre (WUR) > Laboratory of Bioinformatics > Transitorium (building 312) room 1034 > > Dreijenlaan 3 > 6703 HA Wageningen > The Netherlands > > phone: +31 (0)317-483 060 > mobile: +31 (0)6-143 66 783 > e-mail: pieter.neerincx at gmail.com > skype: pieter.online > ------------------------------------------------------------ > > > _______________________________________________ > moby-l mailing list > moby-l at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/moby-l > From vmgcastellanos at yahoo.com Fri Jul 18 17:43:59 2008 From: vmgcastellanos at yahoo.com (Victor Manuel) Date: Fri, 18 Jul 2008 14:43:59 -0700 (PDT) Subject: [MOBY-l] Services inputs and outputs Message-ID: <826306.71959.qm@web57002.mail.re3.yahoo.com> thanks Peter. I would like to use the moby in a different context, but as far as I understand the moby does not offer any service discovery capability unless u follow moby standards; so it works as a kind of yellow pages where u can only be found if you publish according to pre-defined standards, am I right? is there any experience of using biomoby in a non bio setting? Also, from where could I download the bio moby ontology of data types? how are the ontologies being aligned, Mark mentioned that my grid ontologies were being aligned with biomoby ontologies, where can I get the mapped ontology? or for that matter the file in which this alignment is specified. Where can I get the doc describing all the moby standards? Cheers. sorry bout my newbe questions but I am trying to figure out which service discovery mechanism could I use in a non bio setting. Cheers. ----- Original Message ---- From: Pieter Neerincx To: Victor Manuel Cc: moby-l at lists.open-bio.org Sent: Friday, July 18, 2008 9:30:10 AM Subject: Re: [MOBY-l] Services inputs and outputs Hi Victor, On 18?Jul?2008, at 3:52 PM, Victor Manuel wrote: > Is there an ontology describing biomoby inputs and outputs? Yes, that is what BioMoby was designed for in the first place :) > does biomoby provide any web service discovery mechanism? Yes, that is what the BioMoby ontologies mentioned above were designed for in the first place :) > does biomoby standardizes the description of web services in bionfo? Partially, that is what BioMoby was designed for in the first place :) The inputs and outputs are standardised as well as service types, but there is also a "free text form" description field to describe a service. The latter is just meant to be a human readable string and is not standardised (not even on language, so you might find some non- english stuff in there...) Both the inputs & outputs as well as service types and the services themselves are stored in ontologies. These are available from repositories called BioMoby Central. There is one official public one and several institutes host their own BioMoby Centrals either for development or for services which are not publicly available. Looks like you could use a BioMoby primer; the BioMoby 1.0 paper would be good start: http://bib.oxfordjournals.org/cgi/content/abstract/9/3/220 Cheers, Pi > > > > - >> > > _______________________________________________ > moby-l mailing list > moby-l at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/moby-l > > > > > _______________________________________________ > moby-l mailing list > moby-l at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/moby-l ------------------------------------------------------------- Wageningen University and Research centre (WUR) Laboratory of Bioinformatics Transitorium (building 312) room 1034 Dreijenlaan 3 6703 HA Wageningen The Netherlands phone: +31 (0)317-483 060 mobile: +31 (0)6-143 66 783 e-mail: pieter.neerincx at gmail.com skype: pieter.online ------------------------------------------------------------ From markw at illuminae.com Fri Jul 18 10:59:09 2008 From: markw at illuminae.com (Mark Wilkinson) Date: Fri, 18 Jul 2008 07:59:09 -0700 Subject: [MOBY-l] Services inputs and outputs In-Reply-To: <437685.19667.qm@web57005.mail.re3.yahoo.com> References: <437685.19667.qm@web57005.mail.re3.yahoo.com> Message-ID: Yes, yes, and yes :-) ("standardize the description of web services" - Moby's web services descriptions are aligned with those of the myGrid project - we use the same model for annotating a service) Mark On Fri, 18 Jul 2008 06:52:04 -0700, Victor Manuel wrote: > Is there an ontology describing biomoby inputs and outputs? does biomoby > provide any web service discovery mechanism? does biomoby standardizes > the description of web services in bionfo? > > > > - >> > > _______________________________________________ > moby-l mailing list > moby-l at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/moby-l > > > > _______________________________________________ > moby-l mailing list > moby-l at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/moby-l > From pieter.neerincx at gmail.com Mon Jul 21 06:51:05 2008 From: pieter.neerincx at gmail.com (Pieter Neerincx) Date: Mon, 21 Jul 2008 12:51:05 +0200 Subject: [MOBY-l] Services inputs and outputs In-Reply-To: <826306.71959.qm@web57002.mail.re3.yahoo.com> References: <826306.71959.qm@web57002.mail.re3.yahoo.com> Message-ID: <90D55A22-0098-43D4-937F-BBB5976467D5@gmail.com> Hi Victor, On 18 Jul 2008, at 23:43, Victor Manuel wrote: > thanks Peter. I would like to use the moby in a different context, > but as far as I understand the moby does not offer any service > discovery capability unless u follow moby standards; Yes, in order to make automated service discovery by computers possible you'd have to stick to some standard. If there wouldn't be a standard of some kind you could use text mining on human readable descriptions of services, but with the current state of the art in text mining your succes rate in discovering compatible services would be ... eh rather unimpressive. > so it works as a kind of yellow pages where u can only be found if > you publish according to pre-defined standards, am I right? Yes. > is there any experience of using biomoby in a non bio setting? AFAIK You would the first to boldly go were only bioinformaticians have gone before... But I remember a recent discussion on languages used in the ontologies. For bioinformatics supporting English only would be sufficient as it is the de facto standard in Science, but mentioned he wanted to keep the option open to use Moby for other things than Bio... > Also, from where could I download the bio moby ontology of data types? There is not really a web page where you can download the ontologies. BioMoby Centrals are a combination of MySQL databases + Apache web servers and normally you use the BioMoby code base (Perl or Java) to interact with BioMoby Central. Just as a quick hack you could try several URLs like these: http://moby.ucalgary.ca/RESOURCES/MOBY-S/Objects http://moby.ucalgary.ca/RESOURCES/MOBY-S/ServiceInstances http://moby.ucalgary.ca/RESOURCES/MOBY-S/Namespaces These URLs are mostly used by clients who want to download the entire respective ontologies for caching purposes. There should also be a ../ServiceTypes ../Full but nobody is caching that yet and therefore it's not implemented yet. So for that info you'd really have to use the codebase to interact with BioMoby. > how are the ontologies being aligned, Mark mentioned that my grid > ontologies were being aligned with biomoby ontologies, where can I > get the mapped ontology? or for that matter the file in which this > alignment is specified. Maybe Mark can chime in on this one... > Where can I get the doc describing all the moby standards? Cheers. As mentioned before the best primer to get you started would be the the BioMoby 1.0 paper: http://bib.oxfordjournals.org/cgi/content/abstract/9/3/220 Cheers, Pi > sorry bout my newbe questions but I am trying to figure out which > service discovery mechanism could I use in a non bio setting. Cheers. > > > > ----- Original Message ---- > From: Pieter Neerincx > To: Victor Manuel > Cc: moby-l at lists.open-bio.org > Sent: Friday, July 18, 2008 9:30:10 AM > Subject: Re: [MOBY-l] Services inputs and outputs > > Hi Victor, > > On 18?Jul?2008, at 3:52 PM, Victor Manuel wrote: > >> Is there an ontology describing biomoby inputs and outputs? > > Yes, that is what BioMoby was designed for in the first place :) > >> does biomoby provide any web service discovery mechanism? > > Yes, that is what the BioMoby ontologies mentioned above were designed > for in the first place :) > >> does biomoby standardizes the description of web services in bionfo? > > Partially, that is what BioMoby was designed for in the first place :) > The inputs and outputs are standardised as well as service types, but > there is also a "free text form" description field to describe a > service. The latter is just meant to be a human readable string and is > not standardised (not even on language, so you might find some non- > english stuff in there...) > > Both the inputs & outputs as well as service types and the services > themselves are stored in ontologies. These are available from > repositories called BioMoby Central. There is one official public one > and several institutes host their own BioMoby Centrals either for > development or for services which are not publicly available. > > Looks like you could use a BioMoby primer; the BioMoby 1.0 paper would > be good start: > > http://bib.oxfordjournals.org/cgi/content/abstract/9/3/220 > > Cheers, > > Pi > > >> >> >> >> - >>> >> >> _______________________________________________ >> moby-l mailing list >> moby-l at lists.open-bio.org >> http://lists.open-bio.org/mailman/listinfo/moby-l >> >> >> >> >> _______________________________________________ >> moby-l mailing list >> moby-l at lists.open-bio.org >> http://lists.open-bio.org/mailman/listinfo/moby-l > > ------------------------------------------------------------- > Wageningen University and Research centre (WUR) > Laboratory of Bioinformatics > Transitorium (building 312) room 1034 > > Dreijenlaan 3 > 6703 HA Wageningen > The Netherlands > > phone: +31 (0)317-483 060 > mobile: +31 (0)6-143 66 783 > e-mail: pieter.neerincx at gmail.com > skype: pieter.online > ------------------------------------------------------------ > > > > _______________________________________________ > moby-l mailing list > moby-l at lists.open-bio.org > http://lists.open-bio.org/mailman/listinfo/moby-l ------------------------------------------------------------- Wageningen University and Research centre (WUR) Laboratory of Bioinformatics Transitorium (building 312) room 1034 Dreijenlaan 3 6703 HA Wageningen The Netherlands phone: +31 (0)317-483 060 mobile: +31 (0)6-143 66 783 e-mail: pieter.neerincx at gmail.com skype: pieter.online ------------------------------------------------------------- From markw at illuminae.com Fri Jul 25 23:08:30 2008 From: markw at illuminae.com (Mark) Date: Fri, 25 Jul 2008 20:08:30 -0700 Subject: [MOBY-l] Oh, and a thanks to Martin, Eddie, Wendy, and Andreas! Message-ID: I did most of my troubleshooting using the Java Web Start release of Dashboard! :-) What a HUGE benefit that is now that it's so easily accessible and with it's new functionalities! Thanks guys (and gal)! M -- -- Mark Wilkinson Heart + Lung Research Institute Centre for High Throughput Biology UBC From markw at illuminae.com Fri Jul 25 23:00:37 2008 From: markw at illuminae.com (Mark) Date: Fri, 25 Jul 2008 20:00:37 -0700 Subject: [MOBY-l] Apologies, a hint for MIPS-Service substitutions, and an invitation Message-ID: Hi all! It would appear that, for the past few months, all of the DragonDB moby services have been down... because of a missing "use" statement in my Service daemon! Spank me later... but accept my apologies in the meantime...though I guess I must be the only person who uses them given that nobody complained ;-) LOL! Anyway, several of the Moby demo workflows (e.g. the ones that return images of Snapdragon mutations) will now function again properly! Unfortunately, we've lost *all* of our MIPS services, and many of those are components of the various demo workflows, since the MIPS services have historically been the most reliable of all Moby services. Because I needed them for a course I am teaching on Monday, I've "replaced" two of the dead MIPS services with ~equivalent services that wrap TAIR. The Global_Keyword search against mutant phenotypes is now available from the arabidopsis.org authority (I should probably ask TAIR if they mind me referencing their authority name... but it does wrap their web interface so I thought I should acknowledge them as the authority for that service even though they don't host it!). I am also retrieving sequences based on AGI locus codes from arabidopsis.org rather than from MIPS. Here are the two substitution services: getAGILocusCodes provided by: mips.gsf.de can be replaced by getAGI_LocusCode_by_Keyword provided by: arabidopsis.org getEmblDNASequence provided by: mips.gsf.de can be replaced by getTAIR_SequenceByAGI_LocusCode provided by: arabidopsis.org If you make these substitutions in your own workflows you will likely find that your workflows suddenly come back to life! IF ANYONE NEEDS REPLACEMENT SERVICES FOR THE OTHER CURRENTLY DEAD MIPS SERVICES PLEASE TELL ME and I'll do my best to code them up for you (if possible). It only took a few minutes to wrap TAIR as Moby services, so I'm happy to do this for you if you need it. In addition, MIPS provided an EMBL ID retrieval based on keyword, and that fed-into a MIPS sequence retrieval based on EMBL ID. This pair of services can also be replaced by the two above, since they're just a different namespace/id route to the same data. This all brings me to my final comment - I currently have two summer-students who are working on the problem of automated workflow repair (i.e. finding equivalent services to fix broken workflows). We're going to be inviting everyone to participate in a Moby Service Annotation Jamboree later this summer. It will be a "virtual" jamboree - just come to the website. We'll have an interface there through which you can deeply annotate Moby services (inputs, outputs, underlying resources, service types, authorities, etc.) with the goal of improving service discovery. I encourage all Moby users to participate and improve the quality of our service annotations! At a minimum, I hope that the current service providers will visit the website and annotate their own services :-) Best wishes everyone! Mark -- -- Mark Wilkinson Heart + Lung Research Institute Centre for High Throughput Biology UBC From jason.e.stewart at gmail.com Sat Jul 26 01:23:38 2008 From: jason.e.stewart at gmail.com (Jason Stewart) Date: Sat, 26 Jul 2008 10:53:38 +0530 Subject: [MOBY-l] Apologies, a hint for MIPS-Service substitutions, and an invitation In-Reply-To: References: Message-ID: <41c1ade50807252223y3de29b54ub129f2319b296558@mail.gmail.com> Hey Mark On Sat, Jul 26, 2008 at 8:30 AM, Mark wrote: > This all brings me to my final comment - I currently have two > summer-students who are working on the problem of automated workflow repair > (i.e. finding equivalent services to fix broken workflows). We're going to > be inviting everyone to participate in a Moby Service Annotation Jamboree > later this summer. It will be a "virtual" jamboree - just come to the > website. We'll have an interface there through which you can deeply > annotate Moby services (inputs, outputs, underlying resources, service > types, authorities, etc.) with the goal of improving service discovery. I > encourage all Moby users to participate and improve the quality of our > service annotations! At a minimum, I hope that the current service > providers will visit the website and annotate their own services :-) Hey! Cool idea! That makes distant remote people like me actually feel useful!! Keep us posted. Cheers, jas. From MichaelGerlich at gmx.de Sun Jul 27 05:29:40 2008 From: MichaelGerlich at gmx.de (Michael Gerlich) Date: Sun, 27 Jul 2008 11:29:40 +0200 Subject: [MOBY-l] Usage of Array Types and/or Collection In-Reply-To: References: Message-ID: <8C1450CAEAB0401F9182DC1513159783@MichaelPC> Hi all, I encountered some sort of strange behaviour when working with Biomoby Services trough Dashboard and Taverna. When a serviece consumes an array, then you've got the possibility to add multiple entries to this array via dashboard's simple client and these green "+" signs. The xml output also shows that this array element has several child elements according to your inputs. So the service get's invoked one time with the complete list/array (and collection has been disabled when registering the service). If I try to invoke this service via Taverna, I can also add multiple inputs, but this time the service doesn't get called one time, but e.g. 8 times if your array consisted of 8 entries (that makes 8 single input fields in Taverna). I also can see that the xml output created by Taverna shows 8 different inputs/outputs to the moby service instead of the one dashboard gave. Now, after creating another service that didn't used an array, but just a single field of that array and marked collection, things a vice versa. Taverna allows that the inputs get stacked up as a list and the service is called only one time, while in dashboard through using only one field, there is also only one input without the ability to add more entries. Is this strange or is this behaviour wanted? Also Taverna can parse the different Biomoby inputs/outputs - is there some kind of switch to say "add them all as array to make a single input or call to the service? In addition, could somebody tell me when it's right to choose an array as input and when it's better to activate "collection" when registering a service? Is it appropriate to always check collection when dealing with arrays or some sort of multiple inputs? Best wishes, Michael From martin.senger at gmail.com Sun Jul 27 06:21:58 2008 From: martin.senger at gmail.com (Martin Senger) Date: Sun, 27 Jul 2008 18:21:58 +0800 Subject: [MOBY-l] Fwd: Usage of Array Types and/or Collection In-Reply-To: <4d93f07c0807270321v375177aah8a11cc14c4ab2ea5@mail.gmail.com> References: <8C1450CAEAB0401F9182DC1513159783@MichaelPC> <4d93f07c0807270321v375177aah8a11cc14c4ab2ea5@mail.gmail.com> Message-ID: <4d93f07c0807270321j48b4bd89y34aa1899246e0776@mail.gmail.com> ---------- Forwarded message ---------- From: Martin Senger Date: Sun, Jul 27, 2008 at 6:21 PM Subject: Re: [MOBY-l] Usage of Array Types and/or Collection To: Michael Gerlich When a serviece consumes an array, then you've got the possibility to add > multiple entries to this array via dashboard's simple client and these green > "+" signs. Just to clarify what Dashboard does and does not (I guess Eddie will comment on Taverna): The SimpleClient panel is able to create an input data but it is not able to create an input collection with more than one MobySimple element (it is because it was never done, not because it would be a big problem). [One has to edit the input XML message if she wants to have more elements in an input collection.] However, as you observed, Dashboard can create an array of member elements (if such member is defined as being in a HAS relationship) - that's what the green plus is for. Cheers, Martin -- Martin Senger email: martin.senger at gmail.com,m.senger at cgiar.org skype: martinsenger -- Martin Senger email: martin.senger at gmail.com,m.senger at cgiar.org skype: martinsenger From edward.kawas at gmail.com Mon Jul 28 10:09:35 2008 From: edward.kawas at gmail.com (Edward Kawas) Date: Mon, 28 Jul 2008 07:09:35 -0700 Subject: [MOBY-l] Usage of Array Types and/or Collection In-Reply-To: <8C1450CAEAB0401F9182DC1513159783@MichaelPC> References: <8C1450CAEAB0401F9182DC1513159783@MichaelPC> Message-ID: <488dd320.08b38c0a.2fbc.ffffe325@mx.google.com> Hi Michael, In Taverna, you can tell the service to either execute the service (that consumes a collection) one at a time for each item in the collection or all at once (http://biomoby.open-bio.org/CVS_CONTENT/moby-live/Java/docs/taverna/guide/i ndex.html#faq5). Basically, Taverna breaks apart collections for you if you pass one to a service that doesn't consume one (done in the background). This is intentional so that we can have even more interoperability between moby services regardless of the type of input (simple | collection). People tend to use collections whenever your service requires all the data in the collection to be invoked and not when they want to invoke the service x times per entry in the collection. Thanks, Eddie -----Original Message----- From: moby-l-bounces at lists.open-bio.org [mailto:moby-l-bounces at lists.open-bio.org] On Behalf Of Michael Gerlich Sent: July-27-08 2:30 AM To: moby-l at biomoby.org Subject: [MOBY-l] Usage of Array Types and/or Collection Hi all, I encountered some sort of strange behaviour when working with Biomoby Services trough Dashboard and Taverna. When a serviece consumes an array, then you've got the possibility to add multiple entries to this array via dashboard's simple client and these green "+" signs. The xml output also shows that this array element has several child elements according to your inputs. So the service get's invoked one time with the complete list/array (and collection has been disabled when registering the service). If I try to invoke this service via Taverna, I can also add multiple inputs, but this time the service doesn't get called one time, but e.g. 8 times if your array consisted of 8 entries (that makes 8 single input fields in Taverna). I also can see that the xml output created by Taverna shows 8 different inputs/outputs to the moby service instead of the one dashboard gave. Now, after creating another service that didn't used an array, but just a single field of that array and marked collection, things a vice versa. Taverna allows that the inputs get stacked up as a list and the service is called only one time, while in dashboard through using only one field, there is also only one input without the ability to add more entries. Is this strange or is this behaviour wanted? Also Taverna can parse the different Biomoby inputs/outputs - is there some kind of switch to say "add them all as array to make a single input or call to the service? In addition, could somebody tell me when it's right to choose an array as input and when it's better to activate "collection" when registering a service? Is it appropriate to always check collection when dealing with arrays or some sort of multiple inputs? Best wishes, Michael _______________________________________________ moby-l mailing list moby-l at lists.open-bio.org http://lists.open-bio.org/mailman/listinfo/moby-l From groscurt at mpiz-koeln.mpg.de Wed Jul 30 04:51:58 2008 From: groscurt at mpiz-koeln.mpg.de (Andreas Groscurth) Date: Wed, 30 Jul 2008 10:51:58 +0200 Subject: [MOBY-l] Apologies, a hint for MIPS-Service substitutions, and an invitation In-Reply-To: References: Message-ID: <48902BAE.7000501@mpiz-koeln.mpg.de> Mark wrote: > This all brings me to my final comment - I currently have two > summer-students who are working on the problem of automated workflow > repair (i.e. finding equivalent services to fix broken workflows). > We're going to be inviting everyone to participate in a Moby Service > Annotation Jamboree later this summer. It will be a "virtual" > jamboree - just come to the website. We'll have an interface there > through which you can deeply annotate Moby services (inputs, outputs, > underlying resources, service types, authorities, etc.) with the goal > of improving service discovery. I encourage all Moby users to > participate and improve the quality of our service annotations! At a > minimum, I hope that the current service providers will visit the > website and annotate their own services :-) I was wondering how you store this information ? are they mapped to the services/datatypes etc in the moby central database ? Here at the mpiz we already thought about offering this feature as a permanent service to everyone, so everyone can tag services/datatypes etc. Have you ever considered such kind of things ? I recently modified dashboard and my local moby database to have a prototype for service tagging. Unfortunately i can only offer you some screenshots, but I was wondering if such extension could be worth considering to integrate it in the official moby api ? you can see the first idea at http://bioinfo.mpiz-koeln.mpg.de/tagging2.jpg so my suggestion is to think about a tagging system (via dashboard / web portal), which is permant available and allows everyone to tag moby stuff ?! best andreas -- ===================================================== Dipl. Bioinf. Andreas Groscurth Bioinformatics Software Developer Plant Computational Biology group Max-Planck Institute for plant breeding research Carl-von-Linne Weg 10 50829 Cologne Germany +49(0) 221 5062449 ===================================================== From edward.kawas at gmail.com Wed Jul 30 10:59:10 2008 From: edward.kawas at gmail.com (Edward Kawas) Date: Wed, 30 Jul 2008 07:59:10 -0700 Subject: [MOBY-l] [MOBY-dev] Apologies, a hint for MIPS-Service substitutions, and an invitation In-Reply-To: References: <48902BAE.7000501@mpiz-koeln.mpg.de> Message-ID: <489081c6.22bd720a.2434.ffffe4ae@mx.google.com> I like the dashboard panel idea a lot. Seems like the best place to keep that information. Nice thinking Andreas! Eddie -----Original Message----- From: moby-dev-bounces at lists.open-bio.org [mailto:moby-dev-bounces at lists.open-bio.org] On Behalf Of Mark Wilkinson Sent: July-30-08 7:54 AM To: Core developer announcements; moby-l at lists.open-bio.org Subject: Re: [MOBY-dev] [MOBY-l] Apologies, a hint for MIPS-Service substitutions, and an invitation On Wed, 30 Jul 2008 01:51:58 -0700, Andreas Groscurth wrote: > I was wondering how you store this information ? are they mapped to the > services/datatypes etc in the moby central database ? At the moment we're not entirely sure where to put it. I am loathe to put it in Moby Central as we don't yet know what the quality of the annotation will be (though I guess Moby Central isnt' curated either, so... it probably wouldn't hurt). Moreover, the Moby Central API isn't able to query at the level of detail that we will be capturing these annotations, so there's no compelling reason to put it there. For now, we're putting it in a local RDF store, but we may eventually move it into Freebase (http://www.freebase.com/) to enable direct queries there. we'll see how it goes. > so my suggestion is to think about a tagging system (via dashboard / web > portal), which is permant available and allows everyone to tag moby > stuff ?! Certainly I think we need a dashboard panel that contains the annotation interface (called "ED" - the Entity Describer http://entitydescriber.org; written by our very own Eddie Kawas :-) ). ED is currently a bit too tightly linked to Connotea, but we're in the process of ripping it out and making it more standalone for this Jamboree. Eddie, what do you think of a dashboard ED panel? M -- Mark D Wilkinson, PI Bioinformatics Assistant Professor, Medical Genetics The James Hogg iCAPTURE Centre for Cardiovascular and Pulmonary Research Providence Heart + Lung Institute University of British Columbia - St. Paul's Hospital Vancouver, BC, Canada _______________________________________________ MOBY-dev mailing list MOBY-dev at lists.open-bio.org http://lists.open-bio.org/mailman/listinfo/moby-dev From markw at illuminae.com Wed Jul 30 10:54:23 2008 From: markw at illuminae.com (Mark Wilkinson) Date: Wed, 30 Jul 2008 07:54:23 -0700 Subject: [MOBY-l] Apologies, a hint for MIPS-Service substitutions, and an invitation In-Reply-To: <48902BAE.7000501@mpiz-koeln.mpg.de> References: <48902BAE.7000501@mpiz-koeln.mpg.de> Message-ID: On Wed, 30 Jul 2008 01:51:58 -0700, Andreas Groscurth wrote: > I was wondering how you store this information ? are they mapped to the > services/datatypes etc in the moby central database ? At the moment we're not entirely sure where to put it. I am loathe to put it in Moby Central as we don't yet know what the quality of the annotation will be (though I guess Moby Central isnt' curated either, so... it probably wouldn't hurt). Moreover, the Moby Central API isn't able to query at the level of detail that we will be capturing these annotations, so there's no compelling reason to put it there. For now, we're putting it in a local RDF store, but we may eventually move it into Freebase (http://www.freebase.com/) to enable direct queries there. we'll see how it goes. > so my suggestion is to think about a tagging system (via dashboard / web > portal), which is permant available and allows everyone to tag moby > stuff ?! Certainly I think we need a dashboard panel that contains the annotation interface (called "ED" - the Entity Describer http://entitydescriber.org; written by our very own Eddie Kawas :-) ). ED is currently a bit too tightly linked to Connotea, but we're in the process of ripping it out and making it more standalone for this Jamboree. Eddie, what do you think of a dashboard ED panel? M -- Mark D Wilkinson, PI Bioinformatics Assistant Professor, Medical Genetics The James Hogg iCAPTURE Centre for Cardiovascular and Pulmonary Research Providence Heart + Lung Institute University of British Columbia - St. Paul's Hospital Vancouver, BC, Canada From igoldsmid at semanticdiscoverysystems.com Wed Jul 30 15:43:11 2008 From: igoldsmid at semanticdiscoverysystems.com (Ian Goldsmid) Date: Thu, 31 Jul 2008 07:43:11 +1200 Subject: [MOBY-l] Invitation to The Semantic Discovery System V7 Beta Download Message-ID: <001201c8f27c$80c72360$82556a20$@com> Hi all Hopefully this message will have some relevance to some of you. This is a one off announcement ? I am not intending to post any other notices ? at least for several months. Thanks: -------------------------------- The Semantic Discovery System (SDS) V7 Beta 2 is now available for download and use for Non Technical Users, as well for Application Developers (you will find there are two different license types available, one for End User ?Exploring? and one for Developers ? THE LICENSES ARE EMAILED TO YOU A SHORT WHILE, USUALLY 10-20 MINUTES, ?AFTER YOU HAVE REGISTERED AT THE DOWNLOAD SITE, LOGGED IN AND COMPLETED DOWNLOADING). We?ve tag lined the product: Beyond Business Intelligence, from Analytics to Discovery. We are particularly excited about SDS?s capability to enable Non Technical Users to integrate any number of Excel Spreadsheets. This makes it radically easier to perform analyses, queries and reporting across multiple spreadsheets without the usual cutting and pasting, consolidations of one kind or another, and other productivity defeating manual preparations. WE THINK THIS CAPABILITY IS UNIQUE AND IMMEDIATELY VALUABLE - BUT PLEASE TELL US IF YOU AGREE OR NOT. Connecting to Oracle Databases is also made as simple as it can be, such that multiple Oracle and Excel data sources can then be Semantically Queried from one Semantically Savvy Graphical Interface. SDS innovatively uses the latest W3C Semantic Web technologies to enable you to radically improve the effectiveness and timeliness for capturing intelligence , meaning and knowledge from your disparate Corporate Oracle Databases, Excel Spreadsheets AND Web Services. All data sources connected to SDS are effectively integrated at the Semantic level. There is a short 11 slide introductory presentation here: http://www.semanticbusinessintelligence.com You can download the trial version here http://www.semanticbusinessintelligence.com/download.html When you?ve installed SDS you will find that there are numerous example applications (Models/Ontologies & ?canned? Queries) included. And you can find detailed explanations and instructions on using these by downloading the SDS CookBook here: http://www.semanticbusinessintelligence.com/getting_started.html PLEASE REMEMBER THE SOFTWARE IS STILL IN BETA. SOME OF THE FUNCTIONALITY MAY BE BUGGY ON YOUR PARTICULAR ENVIRONMENT. Please report any problems to: support at semanticdiscoverysystems.com We will respond immediately to any questions or issues. Your general feedback about current and desired functionality will be highly valued. Regards, Ian Ian Goldsmid Founder Semantic Business Intelligence UK: +44 208 816 8379 US West Coast: +1 415 691 6097 US East Coast: +1 (617) 379 2797 Asia Pacific: +64 9?889?0269 http://www.linkedin.com/in/iangoldsmid Skype: igoldsmid: http://www.semanticbusinessintelligence.com igoldsmid at semanticdiscoverysystems.com