1. Oxford University e-Science Centre 1 IB-VRE: Integrative Biology VRE Development of a VRE to Support the Integrative Biology Research Consortium Matthew Dovey (Oxford e-Science Centre) David Gavaghan (Oxford Computing Laboratory) Andrew Simpson (Oxford Computing Laboratory) Damian Mac Randal (CLRC) Matthew Mascord (Oxford e-Science Centre) Geoff Williams (Oxford Computing Laboratory) Clint Sieunarine (Oxford Computing Laboratory)

2. Oxford University e-Science Centre 2 Virtual Research Environment A set of online tools, systems and processes interoperating to facilitate or enhance the research process within and without institutional boundaries Environment for managing the entire research process end to end: –Identification of research area & funding source –Collaboration building & management –Resource discovery –Data acquisition –Collaborative analysis of data (AccessGrid, IM) –Dissemination and curation of results (data & knowledge) –Provision of training to new researchers (VLEs, KOS)

3. Oxford University e-Science Centre 3 Grand Unified Virtual Environment

4. Oxford University e-Science Centre 4 JISC VRE Programme Funding £3M from comprehensive spending review: To develop the infrastructure and tools for collaborative e-research environments. JISC Circular 5/04 Developing Virtual Research Environments –August 2004 –Inviting HE institutions in England to submit funding proposals for projects to build and deploy Virtual Research Environments (VREs), based on currently available tools and frameworks. 15 projects funded in the programme Oxford involved in three spanning multiple communities: –VRE for the Integrative Biology Research Consortium (IB VRE) This will build a Virtual Research Environment for the Integrative Biology project, and investigate exposing GRID based services within a VRE as well as richer user interfaces. –Building a Virtual Research Environment for the Humanities (BVREH) This will investigate the needs and requirements of humanities researchers from a VRE –Sakai VRE Portal Demonstrator (Sakai VRE) This project lead by Lancaster and CCLRC will develop a demonstrator SAKAI based VRE addressing a number of research communities including e-Science and Social Science

5. Oxford University e-Science Centre 5 VRE for the Integrative Biology Research Consortium (IB VRE) Integrative Biology Consortium: –EPSRC-funded Integrative biology tackling two of the UKs biggest killers: Heart Disease & Cancer. –Large-scale in silico simulation experiments. –IB building the Grid infrastructure required to support this post genomic research. –Multi-disciplinary & globally distributed collaboration. –IB consortium: 6 UK (Oxford, Nottingham, Leeds, UCL, Birmingham and Sheffield ), 3 US (Tulane, UCSD, UCLA), Canada (Calgary), & New Zealand (Auckland), CCLRC & IBM –Requires a comprehensive set of collaborative research and research management support tools

6. Oxford University e-Science Centre 6 VRE for the Integrative Biology Research Consortium (IB VRE) –To develop a large-scale Virtual Research Environment (VRE) demonstrator –investigate collaboration frameworks to support the entire research process of a large- scale, international research consortium –The key deliverables from the project will be: a robust integrated VRE demonstrator based on OGCE supporting the work of a major multi- and inter-disciplinary research consortium at the interface between the physical and the life sciences detailed reports on the requirements, specification, design, and implementation stages of the development of a fully-functioning VRE to support this leading international research community detailed evaluation studies based on extensive user- and developer-experience of the use and extension of existing VRE middleware to support large-scale collaborative research. –Three core FTE: one at OUCS, two at OUCL Matthew Mascord (Project Manager) Geoff Williams, Clint Sieunarine (Developers) –Close collaboration with CCLRC and the Integrative Biology project

7. Oxford University e-Science Centre 7 VRE for the Integrative Biology Research Consortium (IB VRE) IBVRE Objectives: –Improve the pace, impact, efficiency of integrative biology research in two of the most important problems in clinical medicine today. –Show the potential for the VRE technology developed in other research areas. –Identify areas for improvement in existing VRE standards and technologies –Increase the take-up and sustainability of community source VRE technologies. –Contribute to the development of the e-Research, virtual learning and research support infrastructure at Oxford University.

8. Oxford University e-Science Centre 8 High Level Architecture IBIB-VRE

9. Oxford University e-Science Centre 9 High Level Architecture Presentation layer –General access (e.g. to pre-developed workflows/experiments) via web portal –Specialised access via workbench or command line (e.g. direct webservices from within Matlab) Front end layer –User interface rendering via WSRP to basic IB functions: Resource Location Job Composition Job Submission Computational Steering Data Management (inputs and results) Visualization Workflow management Business logic –Three sub-strata: coordination/orchestration layer –Workflow management –Experimental process –Provenance capture functional services layer –Aggregation of resources/services –Abstraction API for end users base service layer –Direct access to computational and data resources

10. Oxford University e-Science Centre 10 High Level Architecture Resource layer –Underlying IB resources: Computational modelling data storage systems –Other generic VRE resources published and unpublished information (such as e-journals, e-print archives and raw data archives) project management tools (including project wikis and code repositories) Personal Information Management tools (diaries, contact lists, task management, etc) Conference and messaging management Teaching and learning tools Project management tools Bug tracking and enhancement request tools

11. Oxford University e-Science Centre 11 Requirements – Initial Scenarios Identify papers and expertise in a selected area of research. –e.g. search on publication databases or Google –e.g. notification of new papers and publications based on areas of interest. Identify potential sources of funding for collaborative research. –e.g. signing up to receive notification of possible funding opportunities e.g. www.rdinfo.org.uk. Real time textual communication. –e.g. real time discussion include the sharing of visual collaboration through either collaborative results visualisation or real time video communication Real time video communication –e.g. AccessGRID and Personal AccessGRID Real time visualisation and collaborative steering –e.g. running a joint in-silico experiment allowing several researchers to visualise, manipulate and steer a running model Manage workflows –e.g. reuse of processes enabling colleagues to benefit from previous experience and best practice Managing Publications –e.g. managing multiple inputs for collaborative papers

12. Oxford University e-Science Centre 12 Current Progress – Requirements Gathering Advice on requirements gathering strategies from Dr Marina Jirotka, Director of the Centre for Requirements and Foundations at the Oxford University Computing Laboratory Interviews with key users underway Builds on the requirements analysis already undertaken for the IB project Focusing on the wider research process rather than just the core scientific workflow Currently in exploratory phase.

13. Oxford University e-Science Centre 13 Requirements - Interviews Open-ended/unstructured interviews –Recorded with permission –Greater interviewing skills needed –Easier to probe for more information when required Post Interview –Research process is written up and signed-off by the interviewee –Follow sessions arranged where necessary. Typical interview –Initially ask the interviewee to describe a typical day, focussing on higher level scientific process: e.g. research idea -> funding -> hypothesis -> experiment -> analysis of data -> dissemination -> training scientific workflow: write code -> run simulation -> move data -> visualise data un-related activities - how they are interleaved artifacts used in the workspace e.g. lab books, post-its, todo lists –Investigate collaborative activities (why, how and what): How is knowledge divided across the actors in the collaboration, who does what? Synchronous or asynchronous, how time critical is it? Distributed or co-located Public/private Level of awareness of others required when collaborating Technologies used

14. Oxford University e-Science Centre 14 Now and next steps Preliminary findings: –Important to remember that scientists compete as well as collaborate. –Research ideas/agenda usually kept private until publication Next steps: –IBVRE focus group at project workshop –Video observation of scientists collaborating. –Release initial prototype Dec –Iterative analysis of research processes taking into account use of the IBVRE