i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21

journa l homepage: www. int l .e lsev ierhea l th .com/ journa ls / i jmi

Review

Incorporating collaboratory concepts into informatics insupport of translational interdisciplinarybiomedical research

E. Sally Leea,∗, David W. McDonaldd, Nicholas Andersona, Peter Tarczy-Hornocha,b,c

a Division of Biomedical and Health Informatics, Department of Medical Education and Biomedical Informatics,University of Washington, Seattle, WA, USAb Department of Pediatrics, University of Washington, Seattle, WA, USAc Department of Computer Science and Engineering, University of Washington, Seattle, WA, USAd The Information School, University of Washington, Seattle, WA, USA

a r t i c l e i n f o

Article history:

Received 31 August 2007

Received in revised form

16 June 2008

Accepted 30 June 2008

Keywords:

Collaboration

Biomedical informatics

Computer supported collaborative

work

Collaboratories

Social and technical issues

Bioinformatics

a b s t r a c t

Due to its complex nature, modern biomedical research has become increasingly inter-

disciplinary and collaborative in nature. Although a necessity, interdisciplinary biomedical

collaboration is difficult. There is, however, a growing body of literature on the study and

fostering of collaboration in fields such as computer supported cooperative work (CSCW)

and information science (IS). These studies of collaboration provide insight into how to

potentially alleviate the difficulties of interdisciplinary collaborative research. We, there-

fore, undertook a cross cutting study of science and engineering collaboratories to identify

emergent themes. We review many relevant collaboratory concepts: (a) general collaboratory

concepts across many domains: communication, common workspace and coordination, and

data sharing and management, (b) specific collaboratory concepts of particular biomedical

relevance: data integration and analysis, security structure, metadata and data provenance,

and interoperability and data standards, (c) environmental factors that support collabora-

tories: administrative and management structure, technical support, and available funding

as critical environmental factors, and (d) future considerations for biomedical collaboration:

appropriate training and long-term planning. In our opinion, the collaboratory concepts we

discuss can guide planning and design of future collaborative infrastructure by biomedical

informatics researchers to alleviate some of the difficulties of interdisciplinary biomedical

collaboration.

Published by Elsevier Ireland Ltd

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

∗ Corresponding author at: Box 357240, University of Washington, Seattle, WA 98195-7420, USA. Tel.: +1 206 616 0369; fax: +1 206 543 3461.E-mail address: sallylee@u.washington.edu (E.S. Lee).

1386-5056/$ – see front matter. Published by Elsevier Ireland Ltddoi:10.1016/j.ijmedinf.2008.06.011

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21 11

2.1. History of collaboratories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122.2. Theories on collaboratories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

3. Method: approach to literature review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144. Results: review of collaboratory concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

4.1. Review of general collaboratory concepts relevant to any research domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144.1.1. Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144.1.2. Common workspace and coordination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154.1.3. Data sharing and management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

4.2. Review of collaboratory concepts relevant for collaborative biomedical research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154.2.1. Data integration and analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154.2.2. Security structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164.2.3. Metadata and data provenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164.2.4. Interoperability and data standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

4.3. Review of environmental factors that support collaboratories. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164.3.1. Administrative and management structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164.3.2. Technical support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174.3.3. Available funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

5. Future considerations for biomedical collaboration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175.1. Appropriate training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175.2. Long-term planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

6. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17. . . . .. . . . .

1

Teniparorr[piifpta

torwfdirTw[tc

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. Introduction

he modern biomedical research community faces manyxciting and challenging research problems. The complexature of modern research questions has led to a realization

n the research community that a single lab or a single disci-line often can no longer provide all the necessary expertisend resources to solve these questions [1]. Thus, biomedicalesearch has become increasingly interdisciplinary and collab-rative in nature [1–4]. The National Institute of Health (NIH)ecognized the need for interdisciplinary collaboration in itsoadmap for modern biomedical research in the 21st century5]. An element of this recognition of the need for interdisci-linary collaboration is the growing recognition of biomedical

nformatics as an important part of the foundation to supportnterdisciplinary research. A concrete example of the crucialunction biomedical informatics will play in future interdisci-linary collaboration is described in the executive summary ofhe National Meeting on Enhancing the Discipline of Clinicalnd Translational Sciences [6].

As necessary and valuable as this is recognized to be,he literature shows that interdisciplinary scientific collab-ration is not easily established or maintained for severaleasons. First, the researchers involved typically possess aide range of expertise from multiple disciplines with dif-

erent cultures and social norms with no common ground toraw upon to seamlessly communicate [1]. Second, biomed-

cal research in particular is highly competitive and manyesearchers involved are unwilling to share and trust [7–9].hird, many scientists are reluctant to trust unfamiliar tools

hich often results in lack of adoption of core technologies

10]. Fourth, physical proximity is known to be importanto scientific collaboration due to the necessity of informalommunications in fostering collaborative environment, yet

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

interdisciplinary biomedical researchers often lack proximitywith their collaborators [11,12]. Fifth, there is a general lackof a common infrastructure connecting all disparate systemsand workflows due to disparate locations, disciplines, and lackof adoption [13,14].

There is a growing body of literature in fields such as com-puter supported cooperative work (CSCW) and informationscience (IS) on fostering collaboration that provides insightinto how to potentially alleviate the difficulties of interdisci-plinary collaborative research [15–39]. One prominent conceptthat has been heavily researched in the fields of CSCW andIS, is the “collaboratory” [40]. From collaboratory researchemerged many social and technological factors associatedwith collaboration that are useful to consider for facilitat-ing interdisciplinary biomedical collaboration. For biomedicalinformatics to support biomedical research as it becomesmore collaborative, the necessary social and technical aspectsof support will likely be similar to those identified in collab-oratory research in the broader sense described previously.Thus our opinion is that understanding and learning from theexisting collaboratory research is important.

We undertook this literature review for two key reasons.First, none of the existing collaborative research has addressedissues of collaboratories specifically from the biomedical, clin-ical, and translational research point of view. This raisesquestions about where the important overlaps exist and wherethere are gaps that must be addressed when implement-ing collaboratories for biomedical, clinical, and translationalresearch. Second, the existing collaboratory literature has notthus far included a comprehensive review article that includesthe recurring elements across infrastructure, social and envi-

ronmental issues that are critical to collaboratory. This reviewbrings together the wide and diverse range of collaboratoryresults, some of which might be overlooked by others in theirrush to develop novel technologies. The goal of this review is to

i c a

12 i n t e r n a t i o n a l j o u r n a l o f m e d

identify concepts in the collaboratory research literature thatcan inform biomedical informatics to better support interdis-ciplinary biomedical research.

In this paper, we will provide a systematic definition of acollaboratory and a thorough review of collaboratory concepts.We show how general collaboratory knowledge can be appliedto benefit collaborative biomedical research. Furthermore, weidentify more specific aspects of collaboratories that will beparticularly applicable and needed for informatics systems insupport of collaboration in the biomedical field. In order toachieve these goals, we first identify general concepts that arenecessary in all the collaboratories; then proceed to describespecific collaboratory concepts that in our interpretation aredesirable in biomedical collaboration.

2. Background

2.1. History of collaboratories

Collaboratories are “laboratories without walls” whereresearchers can perform their research independent of timeand location. Here, they can interact with colleagues, accessinstrumentation and information, share data and compu-tational resources [40,41]. Although the need for remotecollaboration has been recognized since the introduction ofthe collaboratory concept in 1989, no infrastructure existedto support such research. Hence, early collaboratories wereinitially focused on developing technology to support remotecollaboration such as tools for communication and sharedaccess to instrumentation. For example, the Worm Commu-nity System (WCS) was one of the first collaboratories to formthat provided a collaborative scientific research system forgeographically dispersed biologists [40,42,43].

Many early collaboratories were formed in engineeringdisciplines since remote access to expensive equipment orinstrumentation drove the creation of these engineeringcollaborations. Two notable examples of this type of collabo-ratory include the Upper Atmospheric Research Collaboratory(UARC) [40,44] and the Environmental Molecular Sciences Lab-oratory (EMSL) [40,45,46]. Many of the collaboratories thatfollowed, such as Biological Collaborative Environment (Bio-CoRE) [47,48] and the Biological Sciences Collaboratory (BSC)focused on building systems to support data access, manage-ment, and analysis [49,50].

The term “collaboratory” historically described simple col-laborations focused on the use of technology to access remoteinstrument or data – the term is no longer widely used inthis sense today. Instead, the concept of collaboratories hasevolved and expanded to large collaborations and consortiumsspanning multiple institutions, covering multiple domains ofscience, and leveraging various collaboratory technologies.The collaboratories are now called by other names, such asconsortium, eScience, Grid, center, core and network. Thenature of collaboratory study has also changed. Initially, tool

building was necessary to enable remote collaboration; how-ever, as the collaboratory technologies matured, the focus ofcollaboratories shifted toward the support of both social andtechnical processes of scientific research as whole.

l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21

2.2. Theories on collaboratories

There is no all-encompassing theoretical framework that hasbeen used to describe collaboratories and the related infor-matics issues. Collaboratories are complex socio-technicalsystems that belie a one-solution-fits-all approach. Indeed,most collaboratories are custom designed, each tailored tomeet certain needs. Yet due to the problems and high costsof one-off developments, nearly two decades of collabora-tory research have sought to identify a common technicaland social environment that would facilitate reuse of exist-ing infrastructure while maximizing the adoption and use.The problems in collaboratory research are therefore brokeninto pieces that can be addressed by the different disciplinaryapproaches and theories rather than the overarching ‘collab-oratoryness’.

Collaboratory research has addressed these different areas:

• Design – What is the best way to engage the design of a newcollaboratory?

• Features – What are the most important technical featuresand how do users interact with or operate those features?

• Sharing – How and when are data or physical equipmentshared?

• Communication – How do participants in the collaborationinteract with each other? Should the interaction be throughthe collaboratory explicitly or through other means?

• Coordination – What mechanisms allow participants in acollaboratory to effectively coordinate their individual activ-ities?

• Social Factors – How do social relationships, both existingand new, influence the perception of the system and thecollaborative effort?

• Environment – How does the organizational and institutionalframe in which the collaboratory and the participants sitinfluence the technology and the collaborations?

While there is no single theory that cuts across thestudy of collaboratories, there are several methodologicalapproaches that have been useful in conceptualizing techni-cal and social issues. For example, many of the collaboratorystudies involved researchers who engaged the design ofcollaboratory infrastructure using Participatory Design (PD)[51–54] and User Centered Design (UCD) methods [55]. By thetime collaboratory research had begun, the “strong” designview that theoretical models of collaboration should guidedesign had largely been abandoned. Prior attempts at the-ory driven designs, like that in The Coordinator or ActionWorkflow, two systems that implemented models of humancollaborative activity based on Speech Act Theory, have metwith much user resistance and failure [56]. Adopting PD andUCD methods allowed early collaboratories to begin address-ing domain specific issues of collaboration. These approachesto design cut across the majority of the collaboratory literatureand are still in use by many collaboratory developers.

There are few other methodological or theoreticalapproaches which cross the majority of the research on col-laboratories. This does not mean there is no commonality, justthat the commonality is driven more by the discipline of the

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21 13

Table 1 – A summary of collaboratory concepts and definitions

Collaboratory features Definition

Generalcollaboratoryconcepts acrossmany domains

Communication Communication-asynchronous A method of communication that enablesresearchers to communicatenon-simultaneously (e.g. e-mail, discussionlist, newsgroup)

Communication-synchronous A method of communication that enablesresearchers to communicate simultaneously(i.e. phone, videoconferencing, whiteboard,chat)

Awareness Awareness is an understanding of theactivities of others, which provides a contextfor your own activity

Common workspace,coordination

Common workspace A physical or a virtual space whereresearchers can interact and work together(i.e. web portals, wiki)

Coordination tools Tools that help manage scheduling andcoordination of various tasks involved incollaboration (i.e. scheduler, calendar)

Data sharing andmanagement

Data management Technologies that effectively help handle (i.e.retrieve, search, access) data created duringresearch

Data sharing Any data produced during research beingshared as well as technologies that areinvolved in sharing

Archive/repository A place where researchers can put their datato store, retrieve, and access

Trust A trust is a belief in the integrity and abilityof others involved in the collaboration, whichfacilitate collaborators to work together

Willingness to share A willingness is an inclination and readinessto be part of collaboration and to share in theprocess of a common research goal

Specificcollaboratoryconcepts ofparticularbiomedicalrelevance

Data integration,analysis

Data integration A mechanism to integrate data that are inmultiple formats and located in differentplaces

Data analysis tools Tools that help analyze data in any way (i.e.visualization, display, statistics)

Security Security, variable access A measure to secure access to a system ordata only to those authorized to use it

Metadata, dataprovenance

Metadata (annotations) Contextual information, descriptions about adataset

Data tracking (data provenance) Tools to help keep track of history of a datasetInteroperability,data standards

Common vocabulary/standards A common set of defined words or formatsfor a data set

Interoperability All the technologies in a collaborationintegrated to work together andinterconnected

Collaboratory features Definition

Environmentalfactors that supportcollaboration

Administration, management Management structure A body of people responsible for managingthe overall structure of collaboration

Technical support Technology support Having personnel that help with technology(i.e. set up new hardware, software support)

Funding Funding Funding for collaboration, incentive

Continuous supportof collaboration

Training Education/training Any tools and activities related to educating,mentoring, and training

Iterativeevaluation

Iterative evaluation Continuous evaluation of needs and barriersthroughout the collaborative process (bothtechnological and social)

User-centered design Designing tools with users in mind (i.e.participatory design, ethnographic study ofworkflow)

Understanding workflow Understanding the overall structure of workof those involved in collaboration

i c a

14 i n t e r n a t i o n a l j o u r n a l o f m e d

researcher and the specific problem under investigation. Onlywithin last few years, the dearth of common theoretical frame-works for designing and studying collaboratories lead to theelaboration of TORC (Theory of Remote Collaboration). TheTORC is a descriptive theory that is based on a set of evalu-ation studies and a survey about the critical aspects of remotecollaboration [57].

3. Method: approach to literature review

For this review, we surveyed the literature in a variety of rel-evant disciplines that study and support collaborations. Wespecifically surveyed the literature in Computer Science, Infor-mation Science, Biomedicine and Social Sciences. The focus ofthe search was in technical, as well as social issues relatedto collaboration that look explicitly at collaboratory litera-ture as well as related literature such as CSCW. The primarysearch sites to access literature were Google Scholar [58] andPubMed [59]. We included in our search to other relevantliterature indices and databases such as: Library Literatureand Information Science, Library and Information ScienceAbstracts (LISA), ISI web of Knowledge (Social Sciences Cita-tion Index), Association for Computing Machines (ACM) andthe Institute for Electrical and Electronics Engineers (IEEE).We used published articles, books, conference proceedings,and grey literature such as dissertations and websites. Theinitial keywords searched were: collaboratory, collaboratories,and biomedical collaboration. Subsequent search was per-formed through citation and reverse citation search as wellas through discussions with colleagues in the area of collabo-rative research.

Due to the heterogeneous nature of literature that cov-ers collaboratories, the research cited in this review includesa wide range of publications from different disciplinary per-spectives. The citations cover formative studies designed tocollect requirements for new collaboratories, participatorydesign studies, systems design and deployment studies, aswell as qualitative summative evaluations of collaboratories inuse. The approach taken by these studies varies from descrip-tions of the systems to analyses of the systems without aformal methodology to qualitative analyses of themes acrosscollections of systems. We restricted our review only to the sci-ence and engineering research collaboratories. Although othertypes of collaboratories such as educational or film collabora-tories exist, the challenges within the science disciplines areunique.

For each selected area of literature, we carefully readand developed a list of concepts mentioned. After this listof concepts was compiled, we sorted and categorized indi-vidual concepts so that they can be described in a simplerway (i.e. grouped into the most frequently mentioned con-cepts or less frequently mentioned but relevant concepts).In each of these sorting and categorization processes, thelist of concepts and the categories were validated with co-investigator in the field of CSCW, and then cross-validated

with another co-investigator who is in the field of biomedi-cal informatics. Because of the heterogeneous nature of thearticles being reviewed, much valuable literature is not cap-tured in the existing frameworks but instead in the primary

l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21

literature describing individual systems. We elected to take aqualitative approach to the literature characterizing and cod-ing the themes addressed in each article to identify criticalresults that cross methodological approaches and disciplinaryboundaries. We took this approach rather than the more tradi-tional approach of reviewing serially the existing frameworksfrom a biomedical informatics perspective because we felt thatqualitatively categorizing the emergent and recurring themeswould present a more synthetic overall view of the existingliterature.

4. Results: review of collaboratory concepts

To provide systematic definition and a thorough review ofcollaboratory concepts, we have compiled a list of collabora-tory concepts (see Table 1). All the collaboratory concepts wehave observed in the literature were organized into: (a) generalconcepts relevant for any collaboration, (b) specific conceptsrelevant for biomedical collaboration, (c) environmental fac-tors that support collaboration, and (d) other collaboratoryconcepts. In next few sections, we will illustrate differentaspects of all of these concepts. For each concept in Table 1,we provide a brief definition of each of these collaboratoryconcepts referenced in the literature.

4.1. Review of general collaboratory concepts relevantto any research domain

We identified three essential general collaboratory conceptsthat are both technological and social in nature. The liter-ature suggests that interdisciplinary collaborations cannotfunction without these collaboratory concepts incorporatedinto their structure. They are: (1) communication, (2) com-mon workspace and coordination, and (3) data sharing andmanagement.

4.1.1. CommunicationCommunication and the exchange of ideas and thoughts isa foundational component of collaborative scientific work.In scientific collaboration, all those involved in collabora-tive research formally and informally work together, discussand analyze their research, and share ideas. These formaland informal communications foster collaborative relation-ship by improving working relationships and maintainingshared knowledge [12]. The importance of informal communi-cation in scientific collaboration, such as chance meetings inthe hallways, is well documented [11,12,60,61]. Furthermore,occasional face-to-face meetings are found to be crucial foreffective use of communications tools at distance [62]. It isimportant to note that one reason remote collaborations oftenfail is partly due to the lack of these informal communications[11].

In interdisciplinary scientific collaboration, the process ofcommunication is tightly coupled with communication tech-nology because researchers are often dispersed in remote

locations. Communication technology may be synchronousor asynchronous. Through synchronous tools such as phone,videoconference, or chat systems, scientists can talk to eachother in real time. Real time communication is important

c a l

bssmtuCun

4Awn[sWcapieoLstf

cdAilissItc(sgotAmssc

4Itacegdas

i n t e r n a t i o n a l j o u r n a l o f m e d i

ecause certain methods of communication such as brain-torming cannot otherwise easily occur. Asynchronous toolsuch as e-mail or discussion lists enable scientists to com-unicate without having to set up a specific meeting time;

hus, the researchers are able to work around their busy sched-le by using asynchronous communication mechanisms.ommunication technology can facilitate both planned andnplanned interactions to promote collaboration and aware-ess of other’s research [1,12,60,63,64].

.1.2. Common workspace and coordinationcommon workspace refers to a place where researchers canork together [65]. It is needed for work coordination, commu-ication, and informal information and knowledge transfer

66]. Historically, scientific research typically happened in aingle lab; thus having a common space was not an issue.ith advent of interdisciplinary collaboration brought on by

omplex scientific problems and funding initiatives, havingcommon space is often no longer possible [2]. Interdisci-

linary collaboration frequently involves scientists in multiplenstitutes and labs which are geographically dispersed. How-ver, scientific research cannot succeed without some typef common workspace where researchers can work together.ack of common physical space is especially problematicince as proximity decreases, communication and coordina-ion decrease, and collaboration has an increased chance ofailing [12].

In order to alleviate the problem of a lack of shared space,ommon workspace and coordination technologies have beeneveloped to give scientists a simulated co-located space.

virtual workspace is where all the collaboration relatednformation (e.g., data, communication or software) are col-ected and structured, and can be accessed by the researchersnvolved to coordinate and work together [67]. Through thishared space technology, researchers can come together,hare data, analyze findings, and have research discussions.n case of collaboration, such technology is typically a sys-em accessible by all the researchers involved in the researchollaboration such as Basic Support for Cooperative WorkBSCW) [67,68] or a wiki [69]. BSCW is “a ‘shared workspace’ystem which supports document upload, event notification,roup management and much more” [70]. A Wiki is “a collab-ratively created and iteratively improved set of web pages,ogether with the software that manages the web pages” [69].lthough it is not a physically co-located workspace, com-on workspace and coordination technology can facilitate a

imulated virtual common space. It is important to note thatuch technology is known to reduce negative impact of remoteollaboration [2].

.1.3. Data sharing and managementn collaborative research, data are often generated from mul-iple individual groups. Each individual group must have anbility to effectively manage data generated as well as theollaboration as whole. Data management technologies helpffectively handle data created during research both in smaller

roup settings and in collaboration as whole. Furthermore,ata generated in collaborative research are often shared andnalyzed by all those involved in the collaboration. Hence, dataharing and management go hand in hand in collaborative

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21 15

research. Data sharing is beneficial for several reasons: (1) aresearcher’s findings can be validated by peers, (2) new anal-yses can be performed on existing data, (3) data can serve asthe basis for new research, and (4) it can prevent unnecessaryduplication of effort [71,72]. Further more, new knowledge andinsight can be obtained from collaborative data sharing thatmight not be discovered examining individual data. For thesereasons, funding agencies such as NIH and NSF have beenincreasingly requiring data sharing as part of their fundinginitiatives [6,73,74].

Data sharing often involves technology that enables scien-tists to exchange data. Such technology could be as simpleas spreadsheets which could be sent over an e-mail, or morecomplicated systems such as a common database repositorywhere all researchers can submit their data. Even if there is amechanism to share data, sharing would not be possible with-out trust and willingness to share. Although data sharing isperceived as beneficial, it is hard for scientists to share datadue to the competitive nature of scientific research and thehistorical culture of not sharing [8]. Researchers often do nottrust others involved in collaboration, thinking that when datais shared, someone might steal their findings [57,72], a mis-trust that stems from highly competitive nature of biomedicalresearch [8,9,72]. When implementing data sharing technol-ogy, such social barriers should be taken into account.

4.2. Review of collaboratory concepts relevant forcollaborative biomedical research

We have identified four specific concepts in collaboratoryresearch that are particularly relevant to biomedical collab-oration. Although they are not concepts applicable to allcollaborations, our interpretation is that they are essentialcollaboratory concepts for collaborations in the biomedicalarena. They are: (1) data integration and analysis, (2) securitystructure, (3) metadata and data provenance, and (4) interop-erability and data standards.

4.2.1. Data integration and analysisModern biomedical research increasingly generates largeamounts of data; however, data without analysis is mean-ingless. For example, an image generated by a microarrayexperiment is useless without relevant statistical analysis toindicate what part of that image is significant [75]. In col-laborative research, the researchers often collectively analyzedata as well as review individually analyzed data. A collab-orative analysis can be defined as “an interactive processof brainstorming where researchers share their individualinterpretations, understanding, and insights, which buildupon one another to form cogent findings and conclusions”[50].

Such collaborative data analysis often involves sophisti-cated statistical or analytic tools as well as a mechanism forthe researchers to tie all the analysis processes together [50].Such integrated analysis systems enable scientists to view allthe data in one place, share analysis tools, and capture notes,

working ideas, and interpretations. One important function ofthis type of analysis system is data integration. Without inte-grating data generated by all the researchers involved in thecollaboration, data cannot be analyzed or reviewed collabora-

i c a

16 i n t e r n a t i o n a l j o u r n a l o f m e d

tively. Data integration is also a concept widely known and isactively being researched in the biomedical field due to adventof data explosion challenges [76,77].

4.2.2. Security structureBiomedical research often involves highly sensitive data suchas individual medical records or genetic information. In orderto protect these sensitive data, a carefully planned securitystructure must be in place. HIPAA mandates biomedical datato be constantly under tight security constraints meetingtheir proposed standards [78]. Funding agencies frequentlyrequire that collaborative biomedical research have a care-fully planned security structure [79]. Security structure isnot only a necessity due to sensitive nature of the data. Itis possible that the mistrust among researchers not want-ing to share can be alleviated by tight security constraints[7].

A tight security structure alone, however, is not enough incollaborative research. In collaborative research, certain dataare more sensitive than others. It is also possible that someparts of research might only be available to a small number ofresearchers within the collaboration. To support all levels ofconstraints, a flexible security structure provides a differen-tial access to a common collaborative system [41,80]. Hence,collaborative security structure at bare minimum involvesauthentication (minimum login/password) and authorization(variable access to different parts of research data and workarea) as well as encryption to ensure communication chan-nels are not weak links as well as audit trails to allow reviewof who accessed what.

4.2.3. Metadata and data provenanceBefore the era of large-scale biomedical collaboration, mostresearch projects occurred within a single discipline, often in asingle laboratory. Resulting data were manageable in size andsimilar in content. The researchers had little difficulty under-standing the data since they had similar backgrounds; usefulwhen establishing a common ground or understanding thecontext of the data. Modern collaborative biomedical research,however, generates large quantities of highly diverse data [71].Furthermore, interdisciplinary collaborative research involvesresearchers from various fields. In translational research forexample, at one end of the spectrum there might be a benchscientist sequencing genes involved in a particular cancer,while on the other end a clinical researcher may be involvedin a clinical trial of a drug for cancer [81]. To facilitate col-laborative sharing and analysis, the collaborative researchtools must bridge the gap of diversity in data and variabilityin researchers’ field of knowledge. Metadata associated withdata sets and data provenance are two ways to bridge thatgap [82].

Metadata provides a contextual description of data [83].Without the context of how research was conducted and datawas analyzed, the data are meaningless [75]. For example,say a bench scientist discovers a possible treatment for can-cer and gives the result to a clinical researcher. The clinical

researcher would find the data impossible to decipher withoutan explanation associated with the data. Metadata providessuch contextual explanation. Data provenance is related tometadata in a sense that it is essentially concerned with the

l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21

history of data [82], the importance being that it functions likea versioning mechanism. Without a detailed data history, theresearchers have no way of determining whether the data isoutdated or still valid. Not having data provenance in collabo-rative research could result in confusion of some researchersusing outdated or invalid data while others are using themost updated data. Data provenance enables scientists tokeep track of the large data sets generated by collaborativeresearch.

4.2.4. Interoperability and data standardsInteroperability is a concept assumed in most collaboratoryresearch, but rarely explicitly mentioned. Most collaboratoryliterature implies interoperability by stating how all soft-ware involved in collaboration must be able to talk to eachother [57] or describes collaboratories as all having a com-mon infrastructure support [40]. Interoperability involves acommon infrastructure that seamlessly integrates technologyfrom all levels of research within collaboration. Interoperabil-ity is difficult to achieve in collaborative biomedical researchdue to the heterogeneity of technologies, data formats andcontent at different levels of research [84]. For example, agenetic sequence of a cancer gene is significantly differentfrom an x-ray image of a tumor. Data standards enable inter-operability by creating a common frame of reference. Theresearchers from diverse backgrounds can converge on thatcommon frame of reference to exchange of data and shareideas [85–87].

4.3. Review of environmental factors that supportcollaboratories

Interdisciplinary collaboration does not exist in a vacuum.Collaborations, in particular, are contained within a largerenvironmental framework such as a university or institutionwhich can have positive or negative effect on the collabora-tive process. As such, surrounding environmental factors areimportant to collaborative research. Many studies of collab-oratories focus on either the technical or the social issueswithin the collaboration itself. Very few studies examine theenvironment surrounding the collaboration (consider [42] asa notable exception). From the literature review of these few[57], three aspects emerged as critical environmental factorsthat could support collaboration: (1) Administrative and Man-agement structure, (2) Technical support, and (3) Availablefunding.

4.3.1. Administrative and management structureThe Administrative and Management structure is composedof those involved in collaboration that oversee, coordinate,resolve conflicts, and make decisions. Small collaborationscan succeed without a designated administrative bodysince only a handful of people are involved. The principalinvestigators involved can essentially function as a man-aging body. Large interdisciplinary collaborations, however,involve a large number of people and it is often difficult to

manage them without a designated administrative body. Inthese cases, the administrative structure ideally supportsthe collaborative process by functioning as a central voicefor everyone so that no one involved feels isolated [57].

c a l

Fa[im

4CTtsrklfsctwtuiwisif

4IcpoTbOirtc[

5c

TtqtararccOfwl

i n t e r n a t i o n a l j o u r n a l o f m e d i

urthermore, administrative and management structure canlso function as an infrastructure to support legal issues88]. To properly support anything other than very smallnterdisciplinary collaborations, some body of people has to

anage the overall collaborative process [57].

.3.2. Technical supportollaborative research often involves complex technology.he researchers not only work with remote communica-

ion technologies, but also with tools involved in dataharing and analysis. Since the primary objective of theesearchers involved in collaboration is to advance scientificnowledge, they are often less inclined to dedicate time to

earning new complex technologies. It is possible that therustration that comes with such extra work might detercientists from continuing with the collaboration. Severalollaboratories have failed due to such technical complica-ions. For example, the Worm Community System (WCS)as not adopted by some researchers because they had

o master relatively complex system installations withinnfamiliar computing environments (e.g. the UNIX operat-

ng system) [40]. The upper atmospheric collaboratory usersere challenged by frequent browser downloads [40]. Hence,

nterdisciplinary collaboration can benefit by having technicalupport personnel to alleviate this simple problem of difficultyn technology that could potentially cause collaboration toail [57].

.3.3. Available fundingnterdisciplinary collaboration has more financial overheadompared to traditional research to support overall scientificrocess and technical infrastructure required for coordinationf a large group of people, and sharing of data and analysis.o alleviate this difficulty, adequate funding structure shoulde available as part of the environment fostering collaboration.lson et al. have found that the collaboration based on funding

nitiatives are more likely to fail since the lasting collaborationequires something more than a simple financial incen-ive; however, it would also be difficult for interdisciplinaryollaboration to happen without the incentive of funding57].

. Future considerations for biomedicalollaboration

here is a general consensus that interdisciplinary collabora-ion is important in solving large-scale, complex biomedicaluestions. Since the early days of the collaboratory concept,echnologies such as communication, and data sharing andnalysis tools have been produced to alleviate the problem ofemote collaboration, and social issues of collaboration suchs willingness to share and trust have been identified andesearched. Yet very little is known about how to sustain suchollaborations over the longer term. Over the years, someollaboration simply ended when the initial funding ended.

thers dissolved due to the overwhelming challenges they

aced during the first few years. From our literature review,e identified two factors that might help foster ongoing col-

aboration over multiple years.

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21 17

5.1. Appropriate training

The first factor is appropriate training for both the scien-tific process and technical aspects of collaboration for thoseinvolved in the collaborative research. Because technology isimportant in the daily-to-day functions of an interdisciplinarycollaboratory, the difficulties in technology use could nega-tively effect the resulting collaboration. Additional trainingin technology for technical support personnel can allevi-ate the difficulties in technology adoption and use [57]. Theknowledge of scientific process itself is also important tocollaboration. Due to variability in their background, theresearchers involved in collaboration often focus on their localresearch objectives and overlook the larger goals of the col-laborative effort. It is important for all those involved to havean understanding of the overall goal and the high level scien-tific processes of the entire collaboration. Training in both thetechnology and scientific process, can help foster collabora-tive environment through easier understanding of the overallcollaboration process and use of technology [57].

5.2. Long-term planning

A second factor important in lasting collaboration is long-term strategic planning. Most interdisciplinary collaborationsform out of necessity, either due to a need for expertise or anaccess to funding resources; hence, these collaborations oftenform dynamically, as needs arise. Due to their dynamic nature,collaborative efforts seldom plan beyond the first few years,especially not beyond the initial funding period, althoughthe complex problems that these collaborations seek to solveoften cannot be expected to be resolved in a few years. Long-term planning of infrastructure and the overall collaborativeprocess is needed from the beginning if the collaborations areto last [57]. Lack of such planning has contributed to the end-ing of collaboratories before their aims have been achieved[57]. Furthermore, the interdisciplinary collaborative researchis a constantly changing process. Supporting collaborativeresearch through continuous evaluation of needs, identify-ing gaps/barriers, and learning from past successes/failuresis imperative for the success of long-term collaboration[61,89,90].

6. Discussion

Although recognized by funding agencies such as NIH as anecessity, interdisciplinary biomedical collaboration is diffi-cult for several reasons. Collaborative biomedical researchinvolves a wide range of expertise from different institu-tions in disparate locations with little common ground suchas a common vocabulary, social norms, or organizationalstructure. Based on lessons from collaboratories in other dis-ciplines, the implementation of collaborative technology inbiomedical research will also need to address this lack of com-monality. Furthermore, biomedical collaborations are more

likely to fail due to the highly competitive nature of biomedicalresearch [57].

We have introduced many relevant collaboratory conceptsin this systematic review. Communication, common workspace

i c a l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21

Summary pointsWhat was already known in topic:

• The technologies that enable remote collabora-tion were designed and studied by collaboratoryresearchers in the fields such as Computer SupportedCooperative Work (CSCW) and Information Science(IS).

• The methods to design and evaluate the infrastruc-ture for collaboration were studied in the fields suchas CSCW and IS.

What this study added to our knowledge:

• This study brings in some of the prior collaborationresearches in other fields such as CSCW and IS intothe field of Biomedical Informatics

• This study is a comprehensive review of recurring ele-ments across infrastructure, social and environmentalissues that are critical collaborations. Thus far, noexisting literature incorporates all the elements into

r

18 i n t e r n a t i o n a l j o u r n a l o f m e d

and coordination, and data sharing and management are generalcollaboratory concepts essential to biomedical collaborationwhich all have a biomedical informatics component. Dataintegration and analysis, security structure, metadata and dataprovenance, and interoperability and data standards are specificcollaboratory concepts particularly relevant to biomedicalresearch that are also key areas of research, developmentand application in biomedical informatics. We identifiedadministrative and management structure, technical support, andavailable funding as three critical environmental factors thatcould support collaboration. These factors are overarchingissues important to all aspects of biomedical research col-laborations including the biomedical informatics aspects. Wealso identified appropriate training for both the scientific processand technical aspects of collaboration, and long-term planning astwo factors that can continuously support interdisciplinarycollaboration.

The collaboratory concepts we reviewed provide ways toguide planning and design of collaborative infrastructure tobetter support interdisciplinary biomedical collaboration. Inparticular they have implications for biomedical informat-ics researchers and developers who are working on waysto support biomedical collaborations. The relevance of theseconcepts are illustrated in the latest call for proposal of Clin-ical and Translational Center where it states that biomedicalinformatics support should consider the following: interoper-ability, security, workflow, usability and standards [79]. Theseare only a subset of the concepts we have reviewed. Inthis paper, we have further extended the list of importantconcepts biomedical informatics must consider in order tosupport interdisciplinary collaborative research. It is possiblethat there needs to be an in-depth study of the work prac-tices of those involved in collaboration as well as technologiesinvolved at each level.

We do not claim that the collaboratory concepts area panacea for fostering interdisciplinary biomedical collab-oration nor for developing useful informatics systems tosupport it. The goal of this review is not to resolve all dif-ficulties of interdisciplinary biomedical collaboration, but toprovide concepts that can help better support interdisciplinarybiomedical research collaboration. Furthermore, on the basisof the findings from this review, it is clear that there arebenefits for the biomedical informatics community to getinvolved in collaborative research at every step, particularlyearly in the process. From the initial formation, biomedicalinformatics should be part of planning the implementa-tion of collaborative technology to fit the social processesof collaboration. Throughout the collaborative process, infor-matics support should continuously evaluate and makeappropriate adjustments with the ever changing needs andtechnologies.

In this review, we have focused our efforts on the nationalscience and engineering collaborations; yet interdisciplinarybiomedical collaborations that cross international boundariesare beginning to emerge [91]. A whole new set of issues ariseswhen the collaboration crosses national boundaries. These are

issues in addition to those we have identified here. Hence thefuture direction to fostering biomedical collaboration shouldlook towards the interdisciplinary collaboration at the inter-national level.

a single review.

Acknowledgements

The authors would like to thank and acknowledge NIH NLMTraining Grant (5 T15 LM007442-05) and NIH CTSA Grant (08URR 025014A) for providing the funding to support parts of thiswork.

e f e r e n c e s

[1] N. Hara, P. Solomon, S.L. Kim, D.H. Sonnenwald, AnEmerging view of scientific collaboration: scientists’perspectives on collaboration and factors that impactcollaboration, Journal of the American Society forInformation Science and Technology 54 (10) (2003) 952–965.

[2] J.N. Cummings, S. Kiesler, Collaborative research acrossdisciplinary and institu-tional boundaries, Social Studies inScience 35 (2005) 703–722.

[3] A.R. Horwitz, N. Watson, J.T. Parsons, Breaking barriersthrough collaboration: the example of the cell migrationconsortium, Genome Biology 3 (11) (2002) (pcomment2011).

[4] W. Shrum, I. Chompalov, J. Genuth, Trust, conflict andperformance in scientific collaborations, Social Studies ofScience 31 (5) (2001) 681–730.

[5] National Institute of Health Roadmap, http://www.nihroadmap.nih.gov.

[6] National Meeting on Enhancing the Discipline of Clinical andTranslational Sciences, http://www.ncrr.nih.gov/ncrrprog/roadmap/Executive Summary.pdf.

[7] J.P. Birnholtz, M.J. Bietz, When Do Researchers Collaborate?Toward a Model of collaboration Propensity in Science and

Engineering Research, University of Michigan, Ann Arbor,MI, 2005.

[8] E.G. Campbell, B.R. Clarridge, M. Gokhale, L. Birenbaum, S.Hilgartner, N.A. Holtzman, D. Blumenthal, Data withholdingin academic medicine: characteristics of faculty denied

c a l

i n t e r n a t i o n a l j o u r n a l o f m e d i

access to research results and biomaterials, Research Policy29 (2002) 303–312.

[9] G.M. Olson, S. Teasley, M.J. Bietz, D.L. Cogburn,Collaboratories to support distributed science: the exampleof international HIV/AIDS research, in: Proceedings ofSAICSIT 2002, ACM International Conference ProceedingsSerious, Port Elizabeth, South Africa.

[10] V. O’Day, A. Annette, A. Kuchinsky, A. Bouch, When worldscollide: molecular biology as interdisciplinary collaboration,in: Proceedings of the Seventh European Conference onComputer Supported Cooperative Work, Kluwer AcademicPublishers, Bonn, Germany, 2001.

[11] J.D. Herbsleb, A. Mockus, T. Finholt, R.E. Grinter, Distance,dependencies, and delays in a global collaboration, in:Proceedings of Computer Supported Cooperative Work 2000,ACM, New York, NY, US, 2000.

[12] R. Kraut, C. Egido, Patterns of contact and communication inscientific research collaboration, in: Proceedings of the 1988ACM Conference on Computer-Supported Cooperative Work,Portland, OR, USA, 1990.

[13] Commission on Systematic Interoperability: Ending theDocument Game, http://www.endingthedocumentgame.

[14] National Health Information Infrastructure,http://www.aspe.hhs.gov/sp/NHII.

[15] M. Addis, J. Ferris, M. Greenwood, P. Li, D. Marvin, T. Oinn, A.Wipat, Experiences with e-Science work-flow specificationand enactment in bioinformatics, in: Proceedings of UKe-Science All Hands Meeting, Nottingham, UK, 2003.

[16] L.F. Arnstein, G. Borriello, S. Consolvo, C. Hung, J. Su,Labscape: a smart environment for the cell biologylaboratory, IEEE Pervasive Computing Magazine 1 (3) (2002)13–21.

[17] L.F. Arnstein, R. Grimm, C. Hung, J. Hee, A. LaMarca, S.B.Sigurdsson, J. Su, G. Borriello, Systems support forubiquitous computing: a case study of two implementationsof Labscape, in: Proceedings of the First InternationalConference on Pervasive Computing, Springer-Verlag,Germany, 2002.

[18] J.S. Ash, P.N. Gorman, M. Lavelle, T.H. Payne, T.A. Massaro,G.L. Frantz, J.A. Lyman, A cross-site qualitative study ofphysician order entry, Journal of American MedicalInformatics Association 10 (2) (2003) 188–200.

[19] G. Bafoutsou, G. Mentzas, Review and functionalclassification of collaborative systems, International Journalof Information Management 22 (4) (2002) 281–303.

[20] K.S. Baker, Ecological design: an interdisciplinary, interactiveparticipation process in an information environment, in:Proceedings of the workshop on Requirements Capture forCollaboration in e-Science, Edinburgh, UK, 2004.

[21] K.S. Baker, S.J. Jackson, J.R. Wanetick, Strategies supportingheterogeneous data and interdisciplinary collaboration:towards an ocean informatics environment, in: Proceedingsof the 38th Hawaii International Conference on SystemScience, vol. HICSS38, Big Island, HI, US, 2005.

[22] S. Bly, K.M. Keith, P.A. Henline, The work of scientists andthe building of collaboratories, in: Proceedings of the Group97 International Conference on Supporting Group work,1997, General Atomics Report GA-A22619, Phoenix, AZ, US(Preprint).

[23] T.A. Casper, W.M. Meyer, J.M. Moller, Collaboratoryoperations in magnetic fusion scientific research,Interactions 3 (56) (1998) 56–64.

[24] G. Chin, J. Myers, D. Hoyt, Social networks in the virtualscience laboratory, Communications of the ACM 45 (8) (2002)

87–92.

[25] T.A. Finholt, G.M. Olson, From Laboratories toCollaboratories: A New Organizationalform for ScientificCollaboration, 8 (1) (1997) 28–36.

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21 19

[26] T.A. Finholt, Collaboratories: science over the internet, in:Albert H. Teich, Stephen D. Nelson, Stephen J. Lita (Eds.),AAAS Science and Technology Policy Yearbook, AmericanAssociation for the Advancement of Science, Washington,DC, 2002, pp. 339–344.

[27] D. Fisher, P. Dourish, Social and temporalstructures ineveryday collaboration, in: Proceedings of the SIGCHIConference on Human factors in computing Systems,Vienna, Austria, 2004.

[28] I. Foster, J. Vockler, M. Wilde, Y. Zhao, The virtual data grid: anew model and architecture for data-intensive collaboration,in: Proceedings of the First CIDR—Biennial Conference onInnovative DataSystems Research, Asilomar, CA, US,2003.

[29] E. Halkola, Scientific Collaboration and InformationInfrastructures—Information Practices in LIDET Experiment,2004.

[30] P. Jeffrey, Smoothing the waters: observations on the processof cross-disciplinary research collaboration, Social Studiesof Science 33 (2003) 539–562.

[31] J.S. Katz, B.R. Martin, What is research collaboration,Research Policy 26 (1997) 1–18.

[32] G. Melin, Pragmatism and self-organization: researchcollaboration on the individual level, Research Policy 29(2000) 31–40.

[33] J.D. Myers, T.C. Allison, S. Bittner, B. Didier, M. Frenklach,W.H. Green, et al., A collaborative informatics infrastructurefor multi-scale science, in: The Proceedings of theChallenges of Large Applications in DistributedEnvironments (CLADE) Workshop, Honolulu, HI, USA, IEEEComputer Society, 2004.

[34] J.S. Olson, S. Teasley, Groupware in the wild: lessons learnedfrom a year of virtual collocation, in: In Proceedings of the1996 ACM Conference on Computer-Supported CooperativeWork (CSCW’96), ACM Press, 1996.

[35] T. Schleyer, Collaboratories: leveraging informationtechnology for cooperative research, Journal of DentalResearch 80 (6) (2001) 1508–1512.

[36] D.H. Sonnenwald, R.E. Bergquist, K.L. Maglaughlin, E.K. Soo,M.C. Whitton, Designing to support scientific researchacross distances: the nanomanipulator environment, in:D.S.A.M.E.E. Churchill (Ed.), Collaborative VirtualEnvironments, Springer-Verlang, London, 2001.

[37] S. Teasley, S. Wolinsky, Scientific collaborations at adistance, Science 292 (5525) (2001) 2254–2255.

[38] C.S. Wagner, Six case studies of international collaborationin science, Scientometrics 62 (1) (2005) 3–26.

[39] D.H. Sonnenwald, M.C. Whitton, K.L. Maglaughlin,Evaluating a scientific collaboratory: results of a controlledexperiment, ACM Transactions on Computer-HumanInteraction 10 (2) (2003) 150–176.

[40] T.A. Finholt, Collaboratories as a new form of scientificorganization, Economics of Innovation and New Technology12 (1) (2003) 5–25.

[41] R.T. Kouzes, J.D. Myers, W.A. Wulf, Collaboratories: doingscience on the internet, IEEE Computing Practices 29 (1996)40–46.

[42] S.L. Star, K. Ruhleder, Steps towards an ecology ofinfrastructure: complexproblems in design and access forlarge-scale collaborative systems, in: Proceedings of the1994 ACM Conference on Computer Supported CooperativeWork, ACM Press, Chapel Hill, NC, 1994.

[43] Worm Community System, http://www.canis.uiuc.edu/projects/wcs/index.html.

[44] G.M. Olson, D.E. Atkins, R. Clauer, T.A. Finholt, F. Jahanian,T.L. Killeen, A. Prakash, T. Weymouth, The upperatmospheric research collaboratory, Interactions 5 (3) (1998)48–55.

i c a

20 i n t e r n a t i o n a l j o u r n a l o f m e d

[45] Environmental Molecular Sciences Laboratory, http://collaboratory.emsl.pnl.gov.

[46] A. Schur, K.A. Keating, D.A. Payne, T. Valdez, K.R. Yates, J.D.Myers, Collaborative suites for experiment-orientedscientific research, ACM Interactions 3 (5) (1998) 40–47.

[47] M. Bhandarkar, G. Budescu, W. Humphrey, J.A. Izaguirre,S. Izrailev, L.V. Kalé, D. Kosztin, F. Molnar, J.C. Phillips,K. Schulten, BioCoRE: a collaboratory for structural biology,in: Proceedings of the SCS International Conference onWeb-based Modeling and Simulation, San Francisco, CA, US,1999.

[48] Biological Collaborative Environment, http://www.ks.uiuc.edu/Research/biocore, BioCoRE.

[49] Biological Sciences Collaboratory, http://www.pnl.gov/biology, BSC.

[50] G. Chin, C.S. Lansing, Capturing and supporting contexts forscientific data sharing via the biological sciencescollaboratory, in: Proceedings of the 2004 ACM Conferenceon Computer Supported Cooperative Work, Chicago, IL, US,2004.

[51] S. Bødker, K. Grønbæk, M. Kyng, Cooperative Design:Techniques and Experiences from the Scandinavian Scene.Participatory Design: Principles and Practices, LawrenceErlbaum Associates, Hillsdale, NJ, 1993, pp. 157–175.

[52] M.J. Muller, D.M. Wildman, E.A. White, Taxonomy of PDpractices: a brief practitioner’s guide, Communications ofthe ACM 36 (4) (1993) 26–27.

[53] T. Sumner, M. Stolze, in: M.K.a.L. Mathiassen (Ed.), Evolution,Not Revolution: Participatory Design in the Toolbelt Era.Computers and Design in Context, MIT Press, Cambridge,MA, 1997, pp. 1–26.

[54] C. Weng, D.W. McDonald, D. Sparks, J. McCoy, J.H. Gennari,Participatory design of a collaborative clinical trial protocolwriting system, International Journal of Medical Informatics76 (S1) (2006) 245–251.

[55] H. Beyer, K. Holtzblatt, Contextual Design: DefiningCustomer-Centered Systems, Morgan Kaufmann Publishers,1998.

[56] T. Winograd, Categories, disciplines, and socialcoordination, CSCW 2 (1994) 191–197.

[57] J.S. Olson, G.M. Olson, E.C. Hofer, What makes for success inscience and engineering collaboratories? in: Workshop onAdvanced Collaborative Environments, Redmond, WA, US,2005.

[58] Google Scholar, http://scholar.google.com.[59] PubMed Central, http://www.pubmedcentral.nih.gov.[60] J.D. Hollan, S. Stornetta, Beyond being there, in: Proceedings

of‘ACM SIGCHI’92, Monterey, CA, US, 1992.[61] S. LaCoursier, M. Sarkar, Communication and information

needs and barriers: an international collaboration model, in:Medinfo 2004, 2004.

[62] E.H. Shortliffe, V.L. Patel, J.J. Cimino, G.O. Barnett, R.A.Greenes, A study of collaboration among medicalinformatics research laboratories Artificial intelligence inmedicine 12 (2) (1998) 97–123.

[63] P. Dourish, V. Bellotti, Awareness and coordination in sharedworkspaces, in: Proceedings of CSCW’92, ACM Press,Toronto, Canada, 1992.

[64] J.H. Gennari, C. Weng, J. Benedetti, D.W. McDonald,Asynchronous communication among clinical researchers: astudy for systems design, International Journal of MedicalInformatics 72 (10) (2005) 797–807.

[65] L. Bannon, S. Bodker, Constructing common informationspace, in: Proceedings of the European Conference on

Computer-Supported Cooperative Work ECSCW’97, KluwerAcademic Publishers, Lancaster, UK, 1997.

[66] M.S. Ackerman, D.W. McDonald, Collaborative support forinformal information in collective memory systems,

l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21

Information Systems Frontiers 2 (3–4) (2000) 333–347.

[67] W. Appelt, WWW based collaboration with the BSCWsystem, in: Proceedings of SOFSEM’99, Springer LectureNotes in Computer Science, Milovy, Czech Republic, 1999, p.1725.

[68] W. Hoffelner, Applicabilityof of groupware forcommunication in different project environments—a casestudy, Report from ETH Zürich.

[69] C.S. Wagner, Wiki: a technology for conversationalknowledge management and group collaboration,Communications of the Association for InformationSystems 13 (2004) 265–289.

[70] Basic Support for Cooperative Work, http://www.bscw.fit.fraunhofer.de/, BSCW.

[71] C.A. Ball, G. Sherlock, A. Brazma, Funding high-throughputdata sharing, Nature Biotechnology 22 (9) (2004) 1179–1183.

[72] J.P. Birnholtz, M.J. Bietz, Data at work: supporting sharing inscience and engineering, in: Proceedings of the 2003International ACM SIGGROUP Conference on SupportingGroup Work (GROUP’03), Sanibel Island, FL, US, 2003.

[73] NIH data sharing policy, http://grants.nih.gov/grants/policy/data sharing.

[74] NSF data sharing policy, http://www.nsf.gov/pubs/2001/gc101/gc101rev1.pdf.

[75] D.B. Allison, X. Cui, G.P. Page, M. Sabripou, Microarray dataanalysis: from disarray to consolidation and consensus,Nature Reviews. Genetics 2000 (2006) 55–65.

[76] R. Altman, Share and share alike: a proposed set ofguidelines for both data and software, Biomedical ComputerReview (Summer) (2006) 1.

[77] B. Louie, P. Mork, F. Martin-Sanchez, A. Halevy, P.Tarczy-Hornoch, Data integration and genomic medicine,Journal of Biomedical Informatics 40 (1) (2007) 5–16.

[78] Health Insurance Portability and Accountability Act,http://www.cms.hhs.gov/HIPAAGenInfo.

[79] Clinical and Translational Science Awards RFA,http://www.ncrr.nih.gov/clinicaldiscipline.asp.

[80] B. Blobel, Comparing approaches for advanced e-healthsecurity infrastructures, International Journal of MedicalInformatics 76 (5–6) (2007) 454–459.

[81] S.P. Mankoff, C. Brander, S. Ferrone, F.M. Marincola, Lost intranslation: obstacles to translational medicine, Journal ofTranslational Medicine 2 (1) (2004) 14.

[82] Y. Simmhan, B. Plale, D. Gannon, A survey of dataprovenance in e-science, SIGMOD Record 34 (3) (2005)31–36.

[83] K. Buetow, Cyberinfrastructure:, Empowering a “third way”in biomedical research, Science 308 (5723) (2005) 821–824.

[84] V. Maojo, C.A. Kulikowski, Bioinformatics, Medicalinformatics: collaborations on the road to genomicmedicine? Journal of the American Medical InformaticsAssociation 10 (6) (2003) 515–522.

[85] M. Bruun-Rasmussen, K. Bernstein, C. Chronaki,Collaboration—a new IT-service in the next generation ofregional health care networks, International Journal ofMedical Informatics 70 (2–3) (2003) 205–214.

[86] R. Neches, R. Fikes, T. Finin, T. Gruber, R. Patil, T. Senator, etal., Enabling technology for knowledge sharing, AI Magazine12 (3) (1991) 36–56.

[87] H. Häkkinen, M. Korpela, A participatory assessment of ISintegration needs in maternity clinics using activity theory,International Journal of Medical Informatics 76 (11–12) (2007)843–849.

[88] P. David, M. Spence, Towards institutional infrastructures fore-Science: the scope of the challenge, 2003, Final Report ofthe Oxford Internet Institute project on “The InstitutionalInfrastructure of e-Science: The Scope of the Issues.”.

c a l

i n t e r n a t i o n a l j o u r n a l o f m e d i

[89] B. Kaplan, N.T. Shaw, Future directions in evaluationresearch: people, organizational, and social issues, Methodsof Information in Medicine 43 (3) (2004) 215–231.

[90] D.H. Sonnenwald, Expectations for a scientific collaboratory:a case study, in: Proceedings of the 2003 International

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 10–21 21

ACM SIGGROUP, 2003, ACM Press, Sanibel Island, FL,US.

[91] T. Luukkonen, O. Persson, G. Sivertsen, Understandingpatterns of international scientific collaboration, ScienceTechnology & Human Values 17 (1) (1992) 101–126.