Selican Blog
October 14th, 2010

Richard Liman posted a parable about cross-university collaboration.  We thought it illustrates the value of multi-organizational research collaboration and are reproducing it here with Richard’s permission.

“There is an old scientist named Henry. He worked in a dimly lit laboratory in the basement of a classroom building at a well known university in a college town in New England. He has been researching a major breakthrough for 11 years. After many long disappointing nights, he finally has tangible results.

A few years before this major discovery, Angela, a young German graduate student, starts her PhD thesis at a new university in a city an hour from Munich. She comes up with the same concept as Henry. She did not know about his work. She was thinking about the same problem and reaches the same answers. She is not the only one.

A precocious undergraduate in Buenos Aires named Carlos was also thinking about the problem. Carlos gets his PhD and publishes many articles in his native Spanish on this research. While in school he works as a desk clerk at the Intercontinental. A U.S. patent lawyer who stays at the hotel tells him about a website (http://www.gain-online.org ). On it are recently published patent applications filed by universities. The applications often show research before he reads about it in peer reviewed journals. Carlos went to this site often.

One day he sees two U.S. patent applications: one for Henry’s research and one for Angela’s work. He went to the patent lawyer’s URL and found this link: http://www.litmanlaw.com/Rule-99-Prior-Art-Submissions-in-Published-Applications .

Carlos had not applied for a patent. This is unfortunate because he is the first to develop this highly valuable technology. It could have changed his life and given him the recognition and money to have the research career he dreamed about while sitting at the hotel desk.

Angela and Henry never knew about Carlos, but as a result of having the patent lawyer submit his articles within the two month deadline as prior art in their applications, there is a happy ending. Both want to collaborate with him to further develop the technology. The three work together. The synergy of their combined thinking results in a very valuable patent portfolio shared by three universities receiving funding from licensing royalties paid by a global company”.

Richard’s parable uses a patent attorney’s web site as the catalyst that induced collaboration but the parable would hold true for any other catalyst.  For example, a colleague of Carlos may become aware of Henry and Angela’s work and suggest Carlos contacted them or the three might have met at a scientific meeting and realized the synergy between their work.  The point is the power of the three working together is greater than the sum of their individual contributions.  The end result more than compensates for the effort of seeking out collaborators and forging collaborations.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
September 28th, 2010

The GABRIEL Consoritum was organized to identify key elements in the development of asthma in European countries.  Consisting of 164 scientists from over 19 European countries and Canada, the multidisciplinary consortium included expertise in genetics, epidemiology, and immunology.  The workplan included analysis of genetic and environmental interactions and studying the molecular basis for  factors that increase the risk for industrial asthma.  It also included genetic, genomic, and proteomic studies to identify agents that protect against asthma and used genetics, genomics, and proteomics to discover genetic and microbial factors that cause or protect against asthma.  To achieve these goals, the consortium analyzed over 10,000 children and adults with asthma and over 16,000 non-asthmatics.  The results were published in the September 23, 2010 issue of the New England Journal of Medicine.

A project of this size poses major organizational and coordination challenges. To maximize success the project was organized into 9 work packages, one of which was for project management.  Overall management of the consortium was by a steering committee which appointed a project director who appointed project management staff to deal with day-to-day project management activities.  The steering committee was responsible for assuring that milestones were met, deliverables received, and for top level financial management of the consortium.  The committee also managed any conflicts that arose between the partners and helped assure a smooth workflow.

Three scientifc steering groups monitored scientifc workslows and coordination between sites and implemented quality contorl of data.  Internal advisory boards reporting to the steering committee managed strategy and communication between sites.

The GABRIEL project is a good case study showing how the power of team science can solve complex life science problems.  It also underscores the importance of good organization and project management in team science projects.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
September 21st, 2010

We’ve written about the importance of multidisciplinary teams for modern scientific research.   Scientists are often challenged by the team science approach because their their training does not include teamwork skills.  Recognizing this limitation, the Stanford University Clinical and Translational Science Education Center launched a team science training pilot program in October 2009.  The September 10, 2010 issue of Science Careers has an article describing the program.

The pilot program consisted of a four day workshop organized by Hannah Valantine, Professor of Cardiology at the Stanford University School of Medicine and Margaret Neale, Professor of Organizational Behavior at the Stanford Graduate School of Business.  Professor Neale conducted the course.

Ten teams totalling about 70 people participated.  Each team consisted of between two and 15 members.  The teams varied in their composition and area of science and included several clinical teams.   Each day’s format consisted of an interactive team exercise followed by  a lecture and discussion about the purpose of the exercise.  Exercises covered important aspects of team science but used non-scientific exercises to illustrate their points.

Building a High Performance Team

Workshop participants were told they could sit anywhere when they arrived.  Each seat had a handout with detailed facts about a murder mystery.  Participants were unaware that each handout was different.   Solving the mystery required gathering information from other people with different versions of the handout.  After everyone was present, attendees were asked to form teams and solve the mystery.  Most participants failed because they gathered information from others closest to them, usually members of their own research teams rather than seeking those who might have a missing piece of information.  The exercise illustrated the importance of team diversity in solving problems.  Diverse teams that are efficient at solving problems achieve high performance results.

Leveraging a Team’s Expertise

Attendees were given a list of items needed to survive if lost in a desert and asked to rank  in the order of their importance.  Each attendee first completed the list by themselves.  Then attendees were assigned to teams and the teams performed the exercise as a group.  The group scores were compared with individual scores.  In some cases, teams scores were worse than scores of the best individual member.   The exercise illustrated a failure to leverage the expertise of all members can affect outcomes.  Professor Neale explained to attendees that the result would have been different if team members had discussed their past experience that was relevant to the exercise before starting the ranking.

Conflict and Negotiation

Attendees were randomly grouped into three person teams.  Each team member was given a description of a fictional research company and asked to represent it.  The descriptions were written so that some companies were more powerful than others.  Team members had negotiate to allocate a given amount of research dollars between the three companies.   The exercise showed how power and relationships affect negotiations.

The Big Picture

Each participant was asked to play a role of senior manager, manager, worker, or customer.  Workers were grouped into small teams led by senior managers.  All teams formed a company.  The company had to to create a large version of a famous painting within an allotted time and budget.  Each team was responsible for part of the painting.  All parts had to fit when the exercise concluded to the customers’ satisfaction.  The exercise was successful and illustrated a practical application of the previous days’ lessons and teamwork in action.

The feedback for the pilot was generally positive.  Based on the feedback and suggestions of participants, Professors Valentine and Neale will continue the program.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
September 14th, 2010

In a recent commentary in Science Translational Medicine, Mary L. Disis and John T. Slattery argue that team science is critical for translational research.  They state, “There is an increasing realization that “team science” is needed to tackle and conquer the health problems that are plaguing our society. As an example, solving the obesity epidemic most likely requires the integrated interactions of researchers who study lipid metabolism, genetics, and cell growth; endocrinologists; pediatricians; internists; surgeons; exercise physiologists; nutritionists; behavioral researchers; psychologists; and economists—to name just a few types of specialists”.  Three factors can help create and sustain effective teams:

  • Team Leadership – Leaders who motivate, moderate, and can mentor will connect disparate groups;
  • Infrastructure- Disis and Slattery argue that resources to support teams are critical to team science success.  Strategic resource planning should be part of a team science organizational process; and
  • Team Learning – Shared learning enhances trust between teams.  We believe it can also help increase knowledge and information transfer.

In a  follow up interview with Science Translational Medicine, Disis emphasized communication is at the core of team science success.  Another essential component is eliminating egos of team members.  Scientists are trained to value ideas and  their own creativity.   Team dynamics in team science require group interaction and merging of ideas to solve a common problem.  Egos can prevent the merging of ideas.

Dinsis offered advice about how researchers can work to foster a team environment:

  • Give a lab a functional name instead of the name of the lab head or principal investigator – Functional names encourage lab members to take ownership of their work;
  • Find people outside one’s discipline who bring new knowledge to a project – Disis suggests reaching out to people who will not recapitulate an investigators ideas but add value to a project; and
  • Induce institutional changes that foster team science- Universities and companies reward individual achievements.  Fostering team science will require a change in the reward system.

Disis finally argues that the current ways of conducting science in silos hinders progres.  She states,  ”"If the structure is preventing that greatness, disassemble it. Let’s join together and articulate a better structure that does work around team science”.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
September 7th, 2010

A goal in clinical daignostics is creating a lab-on-a-chip.  A lab -on-a-chip integrates several assays on a chip a few square milimeters or centimeters and allows mulitple assays wth very small sample volumes.  Creating a lab-on-a-chip requires interdisciplinary expertise in microfluidics, interfacikng, minaturization,  processing, nx molecular biology.  The research is complex and requires dynamic interactions between interdisciplinary project teams.  The MIRACLE project is an example of a well-organized lab-on-a-chip undertaking.

MIRACLE is an acronym for Magnetic Isolation and moleculaR Analysis of single CircuLating and disseminated tumour cElls on chip.  The goal of the proejct is to develop a lab-on-achip that detects circulating and disseminated tumor cells (CTCs  and DTCs).  CTCs and DTCs are cells released by tumors that criculate in the body that can cause metasteses.  Determining the number of CTCs and DTCs in blood and analyzing their genetic composition may help design treatments for cancer patients.  CTC and DTC analysis may also be useful for following patients during treatement.

Current methods of CTC and DTC detection are labor intensive, costly, and time consuming.  They involve sample processing and cell isolation.  An analysis can take more than one day.  The lab-on-a-chip consortium hopes to develop a less-expensive, more efficient,  and minimally invasive process for detecting CDCs and DTCs.

According to the MIRACLE web site, “MIRACLE aims to develop a fully automated and integrated microsystem providing the genotype (gene expression profile) of CTCs and DTCs starting from clinical samples. This requires the integration and automation of all sample preprocessing steps including the enrichment, counting, electrochemical characterization and genotyping of the cells. This envisaged, fully-automated MIRACLE test would yield decisive results fast and cost-effective, as compared with contemporary diagnostics tests that may take days.”

The MIRACLE proejct is a collaboration between 14 partners assembled from universities, research institutes, and companies throughout Europe.  It is funded by the 7th Framework Programme of the European Commission. The leader/coordinator of the project is the Interuniversity Microelectronics Center (IMEC) in Belgium.  Participants include one university from Spain, one from Norway, and a research institute in Sweden. Most of the partners are companies of varying sizes in Belgium, Germany, the Netherlands, Sweden, and the United Kingdom.  A list of the project partners is shown here.

There are 10 work packages (subprojects ) that are listed here. Seven are technical. The remaining three are: project management; exploitation; and dissemination, training, and outreach. The project is organized so different tasks will be conducted in parallel. The interactive workplan is illustrated here.  Project teams were selected to avoid duplication of effort and provide all necessary elements of the process from sample preparation and detection to a fully operational lab-on-a-chip platform. 

The MIRACLE project is an ambitious undertaking but has several elements needed for success. 

  • Some of the partners worked together on an earlier project.  Teams that previously worked together are more likely to succeed.
  • Project management and coordination is an integral part of the effort.  Teams in team science projects that lack project management and coordination quickly disconnect resulting in lost productivity and duplication of effort.
  • Complex workflows are clearly defined.  Interactions between and among sites is not left to chance.

We believe the MIRACLE project may serve as a model for team science projects and will be watching as it progresses.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
August 31st, 2010

A recent article by Jason Chew on the Seeking Alpha web site argues “Virtual Biotech Are Here to Stay“.   As startup capital becomes scarcer and investors demand capital efficiency, newer biotech companies cannot create all internal capabilities to bring a compound are biopharmaceutical from the laboratory to to clinical trials.  Using a team science approach, a virtual biotech can be started without lab facilities and work with outside groups specializing in all stages of drug development process.  Virtual biotechs  reap a significant reduction in the cost of drug development because they have low fixed costs.  Chew estimates 25% of virtual biotech costs are fixed costs compared to 75% for traditional biotechs.  Virtual biotechs are also more flexible than traditional biotechs.  They can quickly reconfigure project teams.  A relatively small group of individuals in a virtual biotech can work on a larger number of simultaneous projects.

Angiosyn, a virtual biotech company aquired by Pfizer in January 2005, shows virtual biotechs can be successful.  They developed a compound for macular degeneration to the proof of concept stage in three years at a cost of $5 million.

Virtual biotechs are really a team science collaboration.  Like any virtual research collaboration, the key to success is having the right leadership, selecting the right teams, and building relationships and trust between team sites.   Coordination of activities , communication , and data and information sharing between sites are critical to a project’s success.  Collaboration software designed to connect remote science teams will help assure virtual biotech’s success.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
August 24th, 2010

We have been busily demoing Laboratree to prospective users.   Research collaboration project managers who have seen our demo believe it meets important needs.  Laboratree can helpreduce project coordination costs, improve communications between sites, enhance data, and information sharing, and reduce or eliminate time lost because of duplication of effort at two or more collaborating sites.  We can demo Laboratree via Webex and are happy to continue to show it to all interested parties.  Please send an e-mail to info@selican.comwith your contact information and we will get back to you.

We are finisihing our beta product and expect to start our beta test in the fourth quarter of this year.  We have identified research collaborations for the beta group and want to add a few more.  Please contact us at info@selican.com if you are heading a research collaboration and would like to be considered for our beta test.

We are excited to be entering the product development phase where users will start using Laboratree in their collaboration environments.  Our goal is to work as  partners with users to get their feedback as we continue product development to assure we meet their needs.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
August 19th, 2010

We previously wrote a post about open source drug discovery and followed it with two others about the Structural Genomics Consortium and NEWMEDSConsortium. All are  examples of team science. Our post about the Structural Genomics Consortium notedthat people are skeptical that the open source software model will b e widely adopted by the pharmaceutical industry. Skeptics argue that the open source philosophy conflicts with the requirements of drug development.  Open source promotes sharing of information whereas drug development requires patenting and trade secrets.   Despite the skepticism, pharma continues to adopt the open source philosophy for select projects as reported in the July issue of Nature Biotechnology by Stephen Strauss.  (“Pharma Embraces Open Source Models”, Nature Biotechnology 28, 631-634, 2010).

Drug development is an uncertain process in which several thousand compounds must be produced and evaluated to find one that will receive FDA approval.  In a separate article in Nature Reviews Drug Discovery (9,203-214, 2010) Bernard Munos of Eli Lilly estimates the cost of bringing a drug to market, when all failures are included, is $1.8 million.  Incomplete knowledge of basic biology is a key reason for this dilemma.  Aled Edwardsat the University of Toronto believes that no single organization has the resources to gain completely understand basic mammalian physiology.  Thus, the article by Strauss argues, “…the impetus to adopt an open source or open innovation strategy is driven by a need to refocus resources on driving the best compounds through the pipeline and collaborating on those early parts of drug discovery R&D where problems are shared by companies and often, indeed, across the industry”.

Collaborations between industry, university, and government to identify new targets and find biomarkers are becoming more frequent.  Precompetitive collaborations are organized so participants share intellectual property rights  The dilemma drug companies face is casing them to consider deemphasizing intellectual property, at least on precompetitive research (early stage biology), and sharing negative results.   The Structural Genomics Consortium and NEWMEDS consortium, two academic-industrial consortia, have adopted that philosophy.  As an example, Strauss states 200 researchers in the Structural Genomics Consortium,have contributed >20% of new human protein structures in the Protein Data Bank.   That industry is interested in these publicly available structures is evidenced by 20% of Protein Data Bank requests coming from companies, not universities.

Other examples of precompetitive open drug discovery collaboration and data sharing are the Innovative Medicines Initiative, Altzheimers Disease Neruoimaging Initiative, The Biomarkers Consortium, Predictive Safety Testing Consortium, and the Lilly Phenotypic Drug Discovery Initiative.  All are public-private partnerships that  have the attributes of team science.

An open source approach to drug discovery, even precompetitive drug discovery, is not without challenges. One is standardization of methods between participants.  Another is how open is open.  Some companies will only share select data.  Strauss states open source drug discovery requires advanced agreement about which data will be shared.

Some argue that collaborative projects are not really open source unless they have all elements of open source software development.  We would argue one size doesn’t fit all.  The pharmaceutical industry is adopting those open source collaboration in a way that  fits its needs its needs.  Adopting an open source collaboration to fit specific industry industry needs can only can help spur innovation and facilitate team science collaborations between academia and industry.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
August 17th, 2010

The First Annual National VIVO Conference was held at the New York Hall of Science on August 12-13, 2010.  Attended by approximately 200 people, the conference was a forum for those interested in promoting research collaboration.   It introduced the concept and benefits of VIVO to a wide audience.

VIVO is a project funded by a $12.2 million grant from the National Institutes of Health to create software that will establish a national network of scientists.  Based on the VIVO software developed at Cornell University, the enhanced VIVO is being developed to allow scientists to find others doing similar of complementary work.

VIVO is a semantic web application that is pre-populated with detailed profiles of researchers.  Profiles include name, contact information, position, research interests, publications, teaching, service, and professional affiliations.  VIVO also  includes a search functionality for locating people and information within or across organizations.  When finished, it will offer the option of manually entering data or automatically harvesting human resource data from organizational systems, and publication, membership, and other data from public sources.  It will also provide a means of sharing accurate and precise information about researchers interests, activities, and accomplishments,  and will foster team science by allowing users to identify potential collaborators for new teams or to enhance existing teams.

The VIVO project is a virtual collaboration between seven institutions.  Michael Conlon at the University of Florida is the Principle Investigator.  Other participating institutions are Cornell University, Indiana University, Ponce School of Medicine, The Scripps Institute, Washington University, and Weill Cornell Medical College.  The combined team is over 100 people.  Key team members were at the conference.

Mike Conlon and co-chair Kristi Holmes of Washington University opened the conference.  Mike made the point that VIVO is more than software.  It is also a network and community.  VIVO enables researchers to connect and form teams to explore new opportunities.

The keynote speakers were Noshir Contractor of Northwestern University and Jim Hendler of RPI.  Noshir gave the opening keynote.  His  topic was, “Using Web Science to Understand and Enable Research Networks”.  He discussed the importance of sustaining and creating networks and illustraated several expertise identification tools and network analysis tools including C-IKNOW that is being developed at Northwestern University.  He stressed that users who are passionate about collaboration and have a need for a tool like VIVO will be the evangelists that promote adoption.

Jim gave the keynote on the second day of the conference.  His  topic was, “What is the Semantic Web is Really All About”?  He gave a good overview of the semantic web, explanied elements like RDF and ontologies, and gave examples of web sites built using semantic web technologies and that use linked data.

Keynote talks were followed by several series of simultaneous breakout sessions.  We could not attend all sessions and concentrated on those about VIVO implementation and research networking.  One session highlighted other software for research profiles and viewing research networks include Collexis (Elsevier), Digital Vita (University of Pittsburgh), and Catalyst (Harvard).  There is a goal to make these sites and others like them interoperable with VIVO.

Mike Conlon also made the points that VIVO will

  • need a model for data provenance;
  • use linked open data; and
  • have data visualization capabilities;

The VIVO team also discussed challenges to implementing VIVO in another breakout session.  These include getting institutional human resource data, confirming the provenance and quality of data, and outreach within an organization to get users to adopt VIVO.

There was a poster session at the end of the first day with 19 posters.  Most covered various aspects of VIVO and research networking.  We presented a poster titled, “Laboratree – A Web Based Platform for Team Research Collaboration”.  We will be happy to provide a copy of the abstract or Powerpoint file on request.

We enjoyed the conference and found it informative.  We learned a lot about VIVO and believe it will play an important role in team science.  We also believe there are synergies between Laboratree and VIVO and are exploring these further.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati
August 12th, 2010

We are attending the First Annual National VIVO Conference in New York City on August 12-13.  It promises to be an exciting conference bringing together a diverse group of people interested in research collaboration and team science.  We will be presenting a poster titled, “Laboratree – A web based platform for team research collaboration”.  Please stop by if you are also attending.  We’d like to meet you.  For those not attending , we will be contributing to the conference’s  Twitter stream.  The official hashtag is #VIVO10.  We’ll also blog about it afterward.

Share:
  • Print
  • RSS
  • Twitter
  • Facebook
  • Ping.fm
  • Digg
  • Google Bookmarks
  • Blogplay
  • FriendFeed
  • LinkedIn
  • PDF
  • Slashdot
  • StumbleUpon
  • Technorati