Version 29 (modified by anonymous, 16 years ago)

--

BioHackathon 2008

Towards integrated web service in life science with Open Bio* libraries.

Goals

  • What is the best solution for integration of the web services?
    • Is the list of BioMOBY objects is clean enough to treat with the client program? How about the inheritance?
    • Why BioMOBY doesn't use SOAP's complexType instead of the BioMOBY triple (original XML) data structure?
    • Is there any performance bottleneck in the current system of the BioMOBY?
  • How to integrate arising new data structures such as pathways, metabolomics, and proteomics?
  • What can be done with the Open Bio* projects to support interoperability?
  • How to integrate non-MOBY web service providers (Japanese DDBJ, KEGG, PDBj, CBRC services + EBI, NCBI services)

Summary

The number of web service providers in bioinformatics field is increasing year by year. In theory, these services are interoperable and independent from computer langugages. However, each service uses their own definition of the data types and naming convensions for the methods. Moreover, these services are often not usable by specific languages (partly, due to the lack of compliance of the SOAP/WSDL specification in the langauge's library). This situation prevents these services to be popularized for and used in combination by end users as a infrastructure of the daily research. The BioMOBY project has been tried to overcome the circumstances by defining ontologies for data types and methods used in the services. But the set of data types and methods gathered in BioMOBY looks redundant and complicated to generate typical workflows for end users. Besides, some of the major service providers are not included in the BioMOBY framework yet. In DBCLS, we are trying to develop a new proxy server which translates user's request for the appropriate web service provider. The server will also converts the query and result to the simplified/unified data types, say "50 basic bioinformatics data types" (might be a subset of BioMOBY). Anyway, interoperability can be significantly improved if all Bio* libraries supports them and we just need a practical environment for bioinformatics. Thus, we believe gathering all web service providers and Open Bio* library developers to disucuss about the standardization of objects and methods among web services and programming languages must benefit usability and interoperability for the life science in the next era.

  • Utilizing every bioinformatics web services by 4 major computer languages
  • Standardizing the objects and methods in bioinformatics web services
  • Creating compatible classes for sequences, expressions, pathways etc. among Open Bio* libraries
  • Providing various clients to create and execute bioinformatics workflows
  • Connecting seamlessly: Web service - Grid, Remote - Local, SOAP/WSDL - REST, Async - Sync

Topics

List of meeting topics for BioMOBY, Open Bio*, Workflow clients, and web service providers:

BioMOBY

I would like to ask Dr. Mark Wilkinson to give us a lecture for the following objectives:

  • to make every web service providers understand what does the BioMOBY is
  • current status of the BioMOBY project and how the sites integrated by BioMOBY works
  • current status of the semantic MOBY and its practical usages or goals

and I would like to discuss about:

  • what kind of efforts have been made to popularize BioMOBY
    • why all web services are not compatible with BioMOBY yet?
    • what is the strength of BioMOBY compared with non-BioMOBY web services?
    • what kind of improvements can be made from now?
    • which languages can be used to create BioMOBY clients (only Perl and Java for now?)
      • refine the ontology for data types and methods
      • implement BioMOBY client library in every Bio* projects
      • utilize BioMOBY in NCBI, EBI, KEGG, DDBJ etc. or to create a proxy server which converts these services compatible with BioMOBY

Open Bio*

to be written...

Clients (Taverna, G-language etc.)

to be written...

Web Service providers

to be written...

  • How to increase the number of SOAP compliant services (tools/databases)

Venue

DBCLS and CBRC will provide ideal environment for discussions and developments.

We will prepare KML file for Google Earth.

Schedule

  • 2008/2/10 (Sun) - Arrival ( Narita airport, Japan)
  • 2008/2/11 (Mon) - Presentations & discussions (Roppongi Academy Hills)
  • 2008/2/12 (Tue) - Hack, hack, hack (CBRC)
  • 2008/2/13 (Wed) - Hack, hack, hack (CBRC)
  • 2008/2/14 (Thr) - Hack, hack, hack (CBRC)
  • 2008/2/15 (Fri) - Hack, hack, hack (CBRC)
  • 2008/2/16 (Sat) - Depature

Housing

Attendees

If you could recommend any appropriate persons from the following groups, please let me know:

Starting Points

For a complete list of local wiki pages, see TitleIndex.

Attachments