þÿ

[ ]

1st INCF Workshop on Global Portal Services for Neuroscience

September 3-4, 2007
International Neuroinformatics Coordinating Facility Secretariat
Stockholm, Sweden

Authors

Jaap van Pelt and Gordon Shepherd

Scientific Organizer

Gordon Shepherd, Yale University, New Haven, USA

Workshop Participants

Jan G. Bjaalie, INCF Secretariat, Stockholm, Sweden
Chinh Dang, Allen Institute for Brain Research, Seattle, USA
Mark H Ellisman, University of California San Diego, La Jolla, USA
Daniel Gardner, Weill Medical College of Cornell University, New York, USA
Michael F Huerta, National Institute of Mental Health, NIH, Bethesda, USA
Colin D Ingram, University of Newcastle, Newcastle-upon-Tyne, UK
Kathie L Olsen, National Science Foundation, Arlington, USA
Raphael Ritz, INCF Secretariat, Stockholm, Sweden
Andrea Schenone, University of Genova, Genova, Italy
Gordon Shepherd, Yale University, New Haven, USA
Arthur W Toga, University of California Los Angeles, Los Angeles, USA
Shiro Usui, RIKEN BSI, Wako, Japan
David C Van Essen, Washington University, St. Louis, USA
Jaap van Pelt, VU University, Amsterdam, the Netherlands (Rapporteur)
Robert W. Williams, University of Tennessee, Memphis, USA

Workshop observers

Jim Austin, UK
Esther P Gardner, USA
Andreas Herz , Germany
Luciano Milanesi, Italy
Elli Chatzopoulou, INCF Secretariat
Pontus Holm, INCF Secretariat
Ylva Lillberg, INCF Secretariat

Supported by the EU Special Support Action INCF, the INCF Central Fund and the Swedish
Foundation for Strategic Research

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008

[ ]

1. Executive Summary

Coordinating Development of Portals/

Databases

INCF was established by definition to coordinate activities in

neuroinformatics, i.e., coordination and promotion of databas-

ing/sharing, analytical tools and computational models. Portals

were considered to be basic tools not only in neuroinformatics

activities in the field but also for the INCF mission itself. As

the amount of neuroscience data explodes, so do the numbers

of databases and associated portals, and the relations, often

overlapping, between them. It is urgent that these activities be

coordinated to make them accessible and useful to the neuro-

science community. The committee determined that it is essen-

tial that the INCF take a lead role in this coordination.

1. INCF should be proactive in establishing a list of

maintained and developing neuroscience portals/data-

bases. This list should include links to development sites

and descriptions of the objectives of the project. The lev-

el of disclosure of any new development project should

be regulated by the developer.

2. INCF should: i) act as a broker to bring together indi-

viduals with expertise in particular areas of development,

ii) provide guidance on the availability of development re-

sources, and iii) advise on current "best practice", includ-

ing infrastructure recommendations, design processes,

and scalable source control. This "best practice" should

be conducted to enable interoperability via improved and

pervasive ontology, have accessible API/scriptable inter-

face, and have exposed database schemas.

3. INCF should generate a policy statement on open

source and accessibility of projects in development.

4. INCF could provide a repository or library for the

code base, data, etc., of projects no longer in develop-

ment which might be reused in developing new resourc-

es.

5. INCF representatives should be observers at NSF,

NIH, DOD, Creative Commons, UK e-Science All

Hands, and other meetings on core technologies related

to cyberinfrastructure, and pass appropriate information

on to developers.

6. INCF should play a role in identifying and facilitating

the development of unique value-added projects, such as

an international database of neuronal connectivity.

7. INCF should identify and coordinate requirements

(standards) that assist in the acquisition of data, meta-

data, and development of databases.

In addition, there needs to be a standard practice/policy on the

release of statements and information. Methods and routes for

dissemination vary between internal documents within nodes

and those for external consumption. How do you measure/

evaluate success and how are people to be held accountable?

Success measures must be considered in the context of on-go-

ing funding/support from the National Nodes. Annual reports

can fulfill some of these requirements both in the preparation

and its dissemination. This could include summaries of node

activity, which will help with good practice.

Last but not least, opportunities for running undergraduate

courses in neuroinformatics should be considered as well as

including neuroinformatics in medical training. Educational

material could also be made available via the portal.

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008

[ ]

2. Introduction

The goal of the workshop was phrased as to map out exist-

ing portal services for neuroscience, identify their features and

future plans, and outline opportunities for synergistic develop-

ments. The workshop aimed to discuss alternative formats of

future global and integrated portal services.

The workshop covered two full days and followed a program

of talks and discussions. The first day included brief presenta-

tions by each of the participants, identification of several key

areas for discussion and development, and an initial meeting

of workgroups in these areas. Individual presentations fo-

cused on topics of direct or indirect relevance for the over-

all workshop goal and included areas such as: perspectives

on how to organize neuroscience data on a global scale, ex-

amples of large scale portals existing and under develop-

ment, examples of large, individual databases, examples of

networks of databases, and institutional and funding perspec-

tives. The program on the second day included a continua-

tion of discussions within the workgroups and preparations

of conclusions and presentations, followed by plenary reports

of the workgroups, a general discussion, and working on a fi-

nal draft for the recommendations to INCF on coordinating

neuroscience portals and databases for further dissemination.

The workshop was considered to be timely in orienting

the effort toward a new era of global organization of por-

tals and databases for accessing neuroscience information.

An overview of portal facilities presented at the workshop is

given in Appendix A1. The discussions focused mainly on the

role of INCF in global portal services. The outcomes of these

discussions are summarized in the next section.

3. Workshop Discussions

3.1 Role of INCF in global portal services

for neuroscience

Discussions led to the idea of a `portal of portals' as an instru-

ment for INCF to overview available portal services, to make

bridges between them, and to promote and advise in the de-

velopment of new portal services. INCF could thus provide

a portal of portals, with emphasis on coordination (such as

mirroring of sites in the case of heavy traffic, or providing a

unified look and feel of computational environments). In pro-

viding directories of portals and databases, INCF should not

just duplicate what exists but give added value, for instance,

by adding short descriptions and characterizations of databases

that will be helpful to the users.

Discussions resulted also in the view that INCF should adopt

both a supportive role when different initiatives are taken in the

field, and a proactive role in identifying issues and facilitating

new directions.

The committee discussed a wide range of issues that relate to

the role of INCF in coordination and support of global portal

services. A general model for this coordination is summarized

in the next section.

3.2 Defining a portal

Role/Content of portals

The question "What is a portal?" was raised. Is it just a website

with links, or a system that integratively connects databases?

Portals can function (i) as a resource registry/repository (GEO,

boring but necessary), (ii) as a sophisticated search engine

(Google, NCBI, UCSC, NDG, NIF), (iii) as a knowledge gen-

erator/integrator (BIRN, ABA, GeneNetwork), (iv) as a play-

ground and resource bazaar for experts only (Bioconductor for

array data code), (v) as a dynamic encyclopedia (Wikipedia),

(vi) as a local community resource tool (example of ALZfo-

rum, CTC society site). Expectations of a portal were phrased

as getting good answers to standard questions and getting good

answers to questions we didn't even know we could ask. Por-

tals may also function as grid-based e-science environments

with Biolab and CARMEN as typical examples. Participants

agreed that an INCF portal of portals should be in the service

of the Neuroscience Community in the broadest sense, both in

research and in education. As different user communities are

generally not mutually exclusive, neither are portals servicing

these communities. Thus, while showing much complementa-

rity, portals may also have overlapping functionality. It was

noted that a vacuum exists in using databases and models to

generate testable hypotheses.

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008

[ ]

Lessons learned

A significant contribution came from Arthur Toga, who sum-

marized his experiences with a variety of portal projects for

large, multisite imaging consortia (such as The International

Consortium for Brain Mapping (ICBM), The Alzheimer's

Disease Neuroimaging Initiative (ADNI), The Biomedical In-

formatics Research Network (BIRN-Mouse, Morph, F), The

Huntington's Disease Neuroimaging Initiative (HDNI), The

fMRI Data Center (fMRIDC), The NIH MRI Study of Normal

Brain Development (NIHPD), ACE).

Mission The mission of the project should be clearly defined.

An important issue is whether the project's aims are creating

databases or doing science.

Data There is a general consensus about the data to be made

available through the portal.

Composition of the consortium It is important to know the

people participating in the project, and to allow them to decide

their own participation on the basis of expectations, skills, and

complementarity. The required expertise within the consortium

should be present and proven.

Design The portal needs to be designed for longitudinal use.

Technology/tools The technology for data sharing must be in

place and available for the implementation of the portal func-

tions. An upgrade policy is needed in view of future techno-

logical developments in order to guarantee stability in portal

operations. Otherwise, the project follows a moving target,

controlled by technological developments.

Scientific focus The scientific focus of the project needs to

be well defined (which does not always happen to be the case).

Most projects are too broad, with the risk of reducing the im-

pact of the project. However, there are also examples of good

projects with less focus.

Size of the consortium Different rules appear to be needed

to maintain coherence within large or small consortia. Large

consortia appear to raise many issues of competence.

Duration of the project Long-lasting projects will inevita-

bly face many changes in their context and be confronted with

continuous requests for additions, with the risk that the project

loses focus.

License issues License issues need to be defined at the begin-

ning of the project. One has to be aware that `open source' has

many meanings.

Leadership It must be known who is running the show. A

committee may have diverging views.

Funding Funding is essential for continuity of the portal, oth-

erwise the game is over. This issue inevitably becomes perti-

nent at the end of a project and raises many problems.

Sharing It is important to agree on further differentiation of

data and tool sharing with respect to who (the people to share

with), how (the mechanism of sharing), when (the conditions

for sharing), what (the content to be shared), etc. For instance,

the people to share with can be differentiated according to their

function (such as home, guest, member, leader, manager) and

the mechanisms of sharing can also be differentiated (such as

Browse, Download, Upload, Access content, Project info). The

Table below gives an example of sharing differentiation.

3.3 Developing a portal

Global collaboration

An important goal, outlined by Kathie Olsen of NSF, is ex-

panding support of collaboration on a global scale with in-

ternational partners. For instance, INCF may assist groups of

database developers in Japan in finding partners, and give sup-

port for the international embedding of Japanese resources.

Enabling technologies

Portal services are strongly driven by enabling technologies

such as high-performance computing, GRID infrastructures

and visualization tools. Examples of such portals are BIRN,

CARMEN and Biolab. It is strongly recommended that INCF

builds on the opportunities arising from these technological

developments.

Home Guest

Member

Leader

Manager

Browse

Download

Upload

Access content

Project info

Table 1: Example of sharing differentiation

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008

[ ]

Sustainability

Questions were raised concerning the lifetime of data, archiving

the data, and how to maintain and support database facilities

over the long term, requiring high investments in data centers.

These questions essentially concern sustainability issues. The

importance of sustainability was recognized by INCF, which

is organizing a dedicated INCF Workshop on Sustainability of

Neuroscience Databases, December 13-14, 2007, at the INCF

Secretariat in Stockholm, Sweden.

Interoperability

INCF has a primary role in promoting interoperability. Adher-

ence to interoperability requirements may lead to convergence

on common architectures. INCF will build on "best practic-

es" from other experienced web communities such as ATOM

(

http://www.atomenabled.org/

), OAI-PMH (

http://www.

openarchives.org

), RDF (

http://www.w3.org/RDF

), OWL

(

http://www.w3.org/TR/owl-ref

), and OpenID (

http://openid.

net

). Interoperability makes it possible to put findings of dif-

ferent projects together, which may reveal surprising new cor-

relations (an example mentioned by Arthur Toga is that studies

of cortical maturation show that it begins in childhood at the

central sulcus, while independent studies of ageing show that

degeneration occurs in reverse order; database interoperability

brings these two complementary findings together).

Evaluation

The question concerning the organization around portals was

discussed. Japan has adopted a scheme of annual evaluations

of platforms and databases, and this was determined to be a

good model for INCF. Criteria must be formulated for these

evaluations.

Steering committee for portal services

It was agreed that a steering committee is needed for support-

ing this INCF coordination role. The present workgroup on

Portal Services could become such an INCF steering commit-

tee on portal services, acting in the first instance through the

organization of annual workshops.

3.4 INCF role in coordination & support

Many issues of INCF coordination concern data-related con-

cepts, such as data models, data formats, definitions of terms,

ontologies, etc., which are the formalisms that allow compu-

tation upon, and organization and querying of, the data. By

coordinating data-related concepts, informatics tools and in-

formatics resources will become increasingly accessible and

interoperable. It must be recognized that such coordination

will be an ongoing and dynamic process.

When to coordinate?

INCF should carefully select the areas for further stimulation.

The community to address needs to be selected as being best

poised, and be willing to be coordinated and to provide the

required input of data and information. The community needs

to be identified (for instance, by the type of data being used

or the domain of research or the existence of digitized data

or the use of neuroinformatics tools). An educated estimation

of the extent of the task is needed (i.e., how big is the bite?).

It is recommended that INCF formulates a list of criteria that

helps selecting areas for coordination. Also the existence of ad-

jacent communities, where neuroinformatics tools are already

routinely used, is important. It was deemed crucial for INCF

to find and coordinate selected areas for pilot projects, which

make something happen and stimulate new developments.

Initiation of new projects/developments

INCF implements its coordination role by organizing work-

shops, providing portal services, stimulating ontology develop-

ment, and supporting topical program areas such as large-scale

modeling and digital atlasing. The workshops are the instru-

ments to recommend projects and themes for further devel-

opment. Recommendations may also concern new services to

be supported by a portal, or even setting up new portals. Rec-

ommendations for new initiatives, such as building databases,

should be based on actual overviews of current developments

in the areas concerned, such as databases in development.

INCF has a primary role in producing these overviews. INCF

may initiate unique value-added projects, such as facilitating

data mirroring through collaborations. A concrete example is

the collaboration with the Allen Brain Institute to set up a mir-

ror site for the Allen Brain Atlas.

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008

[ ]

Assistance

INCF can provide assistance in the development and matu-

ration of portals and databases, in the formulation of open-

source datasharing/databasing policies, and in the formulation

of criteria for acceptance. Such assistance can be targeted to

developers, to funding agencies and journals, and to industry,

etc. INCF can play a role in periodic evaluations of databases

and providing constructive feedback. INCF can function as an

honest broker to help individual investigators in searching for

the availability of existing databases and databases in develop-

ment for sharing, or in finding alternative data sources. INCF

can provide logistical support. There are potentials for INCF

toward atlas consensus, spatial reconciliation, and directions

for finding data to test datasets for segmentation.

User communities

User communities are the raison d'être of portals and databases.

They define the spectrum of integrated services to be provided

by the portal, such as information about PhD programs, con-

ferences, meetings, etc, and with links to databases and other

portals, to tools and technical developments. For INCF to play

a meaningful role it is crucial to have information about the

communities serviced by the portals, i.e. who are the people or

committees involved in the portals. Services supporting social

interactions are also important, as was demonstrated by Shiro

Usui for Japan, where a variety of tools is already available via

the internet for organizing feedback and user interaction, such

as blogging, tagging, planets, clouds --examples from indus-

try, such as Amazon, can also be cited.

National Nodes

It was discussed whether each National Node should develop

its own portal, and what role INCF should play in their coor-

dination. Because National Nodes have a coordinating role at

the local level, they may wish to develop their own portals to

accommodate services for their "local" communities. INCF

could/should watch over their global embedding.

Technical development of portals

INCF can provide advice and resources in the technical de-

velopment of databases. Assistance can be given to implement

interoperability requirements as well as standards on data and

metadata, ontology "standards" and the use of markup lan-

guages (such as NeuroCommons Mashup Browser in ABA,

for example), the integration in semantic webs, open-source

policies, etc. INCF can also advise on infrastructure, e.g., with

reference to the NSF cyberinfrastructure, and possibly with

support from NSF observers. It was noted that many policies

and standards are currently in various phases of development

themselves, yet that is no reason to wait until their stabiliza-

tion. The graphical toolbox of Shiro Usui could be helpful in

visualizing the relations and homologies in different ontology

and nomenclature schemes. These tools may also help visual-

ize relationships between portals.

3.5 Opportunities

Industry

INCF may coordinate processes toward new standards in in-

strumentation as a service to communities and industry.

Education and training

Portals may additionally be instrumental in the coordinating

role of INCF with respect to education and training in neuro-

informatics. INCF may promote the development of targeted

E-science programs and virtual lab facilities, for instance via

EU funded Initial Training Networks.

Link portals with journals

Transformations in the scientific publication process may dra-

matically impact how neuroscientists share and mine experi-

mental data. A recent PubMed Plus conference in St. Louis,

MO, USA, June 2007, on "New Directions in Publishing and

Data Mining" discussed the formation of a new Neuroscience

Peer Review Consortium for sharing reviews between neuro-

science journals. This Consortium has been operational since

January 1, 2008. INCF could also advise journals by notifying

them when data is suitable for database storage, and which da-

tabases are available for different kinds of data.

4. Recommendations

4.1 Recommendations for a General

Coordination Model and Roles

As the amount of neuroscience data explodes, so do the num-

bers of databases and associated portals. It is urgent that these

activities be coordinated to make them accessible and useful to

the neuroscience community. It is essential that INCF takes a

leading role in this coordination.

What is being coordinated by INCF? Fundamentally, it is data-

related concepts such as data models, data formats, definitions

of terms, ontologies, etc., which are the formalisms that allow

computation upon, and organization and querying of, the data.

By coordinating data-related concepts, informatics tools and

informatics resources will become coordinated and increas-

ingly interoperable. It is recommended that a general model

for this coordination would include the following elements:

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008

[ ]

· Recognition that coordination will be an ongoing and dy-

namic process

· Roles in coordination

- INCF provides organizational and logistical support

- Leadership and implementation comes from the re-

search community

o Thought leaders

o Community-at-large

- Iterative process between products by thought leaders

and modification based on input from community-at-

large

- Implementation by those developing informatics tools

and resources (e.g., databases)

· Parsing neuroscience into subcommunities

- Based on data types (e.g., genetic, circuits, neuroimag-

ing, etc.)

- Identify opportunities to integrate across these (see bul-

let below)

· Keeping in mind potential hooks to informationally-ad-

jacent subcommunities (e.g., genetic data-defined com-

munity and protein structure data-defined community) as

coordination within a subcommunity develops, so con-

ventions and standards that emerge in one community

can be leveraged by, and serve as a potential venue of

integration with, another community

· Identification of which subcommunities are well poised

or ill poised for such coordination--work first with

subcommunity(s) that are best poised

- Develop criteria to identify which is well or not well

poised. Potential criteria might include:

o Are data widely gathered in digital form?

o Do members use informatics and other computational

tools?

o Is there evidence of the need for coordination of infor-

matics efforts (e.g., via editorials, recommendations

from meetings, etc.)

· INCF coordination might be at the level of disseminat-

ing or otherwise facilitating the use of particular, already

generally accepted (if only by major players) data-related

concepts (e.g., data models, data formats, vocabularies,

ontologies) OR at the level of sharing information about

developing and emerging data-related concepts.

4.2 Specific recommendations to the INCF

· INCF should identify relevant portals worldwide and

evaluate their complementarity or overlap. National

Nodes should be encouraged to participate in this analy-

sis. There is an urgent need to coordinate this vast array

of portals and data.

· INCF could serve as an honest broker across neuroin-

formatics efforts. INCF could not only point to ongoing

neuroinformatics tools and resources, but also character-

ize these, using objective criteria, providing this infor-

mation to potential users. Such criteria could relate to

specific aspects (such as ease of data entry and suitability

of particular databases for sharing data in particular com-

munities, etc.).

· The criteria developed to characterize neuroinformat-

ics tools and resources could be posted and the research

community could be encouraged to indicate how well a

particular tool or resource satisfies each criterion.

· INCF could educate investigators about the availability

and utility of neuroinformatics tools and resources use-

ful for their research. INCF might also mediate between

investigators and database managers.

· INCF should develop and sustain expertise about data-

bases suitable to allow data sharing connected to journal

publications.

· INCF could work with journals to facilitate data shar-

ing by increasing awareness of databases appropriate to

their respective journals. Reviewers and editors could

encourage authors to submit their data to relevant data-

bases when appropriate.

· INCF should help investigators to deposit their data into

appropriate databases.

· INCF could serve as a coordinator of policies related to

the sharing of neuroscience data, and neuroinformatics

tools and resources across different funding agencies

(public and private) across different countries--perhaps

by stating best practices and promulgating those across

these agencies and countries. A useful starting point

would be for INCF to ask each node to provide informa-

tion about such policies of major neuroscience and neu-

roinformatics-funding organizations in their country.

· INCF should facilitate training opportunities across in-

formation science and neuroscience at undergraduate

student, graduate student, and postdoctoral levels.

· INCF should establish a steering committee to provide

advice on the implementation of these recommendations.

It is critical to have follow-up that would include physi-

cal meetings of this committee at appropriate venues.

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008

[ 10 ]

Appendix A: Summary of portals discussed during the workshop

URL addresses

Remarks

Japan node and platforms

http://www.neuroinf.jp

http://nijc.brain.riken.jp

Organization and quality - Steering committee,

Platform coordinating committee, annual evaluation,

each platform services own community

NIF

Neuroscience Information

Framework

http://neurogateway.org/

NIH - catalog of electronic and non-electronic

neuroscience research resources,

SumsDB

Surface Management

System Database

http://sumsdb.wustl.edu/sums/

index.jsp

Repository of brain-mapping data

NDG

Neuroscience Database

Gateway

http://ndg.sfn.org/

Selection criteria for inclusion (level of maturation,

functionality, how rapid is information obtained)

SenseLab

http://senselab.med.yale.edu/

Models of neurons and neural systems

Neurondb

The neurophysiology

database

http://neurodb.dertech.com/

Database for neurophysiological data storage,

retrieval, and analysis

BrainPharm

http://senselab.med.yale.edu/

BrainPharm/

Database to support research on drugs for the

treatment of different neurological disorders.

Bio-Lab

Laboratory for bioimages

& bioengineering

http://www.bio.dist.unige.it/

DIST Genova portal function, grid approach,

clinical applications. Integration of data and services,

complexity hidden for users

Neurodatabase.org

http://neurodatabase.org

Neurophysiology database + tools anatomical

visualization

NeuroMorpho.Org

http://neuromorpho.org/

Inventory of digitally reconstructed neurons

ABA

Allen Brain Atlas

http://www.brain-map.org

Image database of gene expression in the mouse

brain.

CARMEN

Code Analysis, Repository

and Modelling for e-

Neuroscience

http://www.carmen.org.uk

EPSRC e-science Pilot Project

Including database for electrophysiology, virtual

laboratory, grid enabled, repositories, toolkit,

integration of electrophysiological and morphological

data

BIRN

Biomedical Informatics

Research Network

http://www.nbirn.net/

A National Information Infrastructure to Enable and

Advance Biomedical Research

CCDB

Cell Centered Database

http://ccdb.ucsd.edu/

High resolution 2D, 3D and 4D data from light and

electron microscopy,

The GeneNetwork

http://www.genenetwork.org

Resources for systems genetics.

NITRC

The Neuroimaging

Informatics Tools and

Resources Clearinghouse

http://www.nitrc.org/

NIH - neuroimaging resources for fMRI and related

structural analyses

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008

[ 11 ]

Appendix 2: Workshop Program

September 3, 2007

08.30 09.00

Introductions (Bjaalie and Shepherd)

09.00 18.00

Scientific presentations and discussions

Gordon Shepherd

Introduction and orientation

David C. van Essen

Opportunities and challenges in navigating

Shiro Usui

Japan-node portal and the platforms

Andrea Schenone

Some experiences in sharing data and services through e-science portals

Gordon Shepherd

Experiences building the Neuroscience Database Gateway

Daniel Gardner

The Neuroscience Information Network: An open source solution and template for web
neuroscience

Chinh Dang

Creating a community-driven neuroscience webware

Jaap van Pelt

Database opportunities for the Netherlands

Colin D. Ingram

CARMEN: Developing an e-science portal to support collaborative research in time-se-
ries neuroscience data

Mark H. Ellisman

The Biomedical Informatics research Network (BIRN) and the Cell Centered Database
(CCDB) Building collaborative environments for integrating and sharing neuroscience
knowledge

Arthur W. Toga

Multi-site imaging consortia: Lessons learned

Robert W. Williams

Interoperability among mouse genomic, neurogenetic, and neuroimaging web resources

Raphael Ritz

The INCF Neuroinformation Community Portal

Michael F Huerta

Neuroscience and informatics: Intersecting priorities at NIH

Kathie Olsen

NSF: Perspectives on cyberinfrastructure and neuroscience

20.00

Dinner

September 4, 2007

08.30 12.00

Discussion

12.00 13.30

Lunch

13.30 18.00

Discussion

Each presentation was approximately 20 minutes, including questions.

Nature Precedings : doi:10.1038/npre.2008.1779.1 : Posted 8 Apr 2008