Training data set: Genomic and clinical data from 2,000 women diagnosed with breast cancer (theMETABRIC data set)
Data access and analysis tools: Synapse
Computer resources: Each participant was provided with a standardized virtual machine donated by Google
Scoring: Models were submitted to Synapse for scoring on a real-time leaderboard
John Wilbanks, the company's Chief Commons Officer, explains:
"This challenge was open source and encouraged code-sharing to forge innovative computational models. The standardized and shared computational infrastructure enabled participants to use code submitted by others in their own model building, and the winning code must be reproducible.
We were using a brand new dataset to select the winning model, derived from approximately 180 breast cancer samples, with data generation funded by Avon. The winning model would be the one that, having been trained using METABRIC data, was most accurate for survival prediction when applied to a brand new dataset.
The winner received, in addition to peer acclaim, the right to submit a pre-approved article about his/her winning model to Science Translational Medicine.
354 participants from 35 countries registered for the challenge. Throughout the model-training period, more than 1,700 models were submitted to the leaderboard on Synapse for scoring (the code for all the models on the leaderboard was accessible to all to use to evolve new models).
Each participant was invited to submit up to 5 of their models, from all of those that they had generated, for final scoring against the METABRIC data set. The winning model was the Attractor Metagene Model, which was 76% accurate against the newly generated OsloVal dataset."
Paper: Crowd Computing: Using Competitive Dynamics to Develop and Refine Highly Predictive Models
From the National Center for Biotechnology website: A recent application of a crowd computing platform to develop highly predictive in silico models for use in the drug discovery process is described. The platform, Kaggle™, exploits a competitive dynamic that results in model optimization as the competition unfolds. Here, this dynamic is described in detail and compared with more-conventional modeling strategies. The complete and full structure of the underlying dataset is disclosed and some thoughts as to the broader utility of such 'gamification' approaches to the field of modeling are offered.
Joerg Bentzien, who attended the Bellagio Center conference, was one of the authors.
Open Source Pharma Letter to WHO to Support an Open Source Ebola Project
To: Dr Marie-Paule Kieny Assistant Director-General, HQ/HIS/HIA World Health Organization
Dear Dr. Kieny,
We understand that you are spearheading “New and Experimental Interventions“ for Ebola and write to you in this capacity.
We have been following the developments of the Ebola outbreak in Africa. We have also been noting calls by several experts on this disease and in public health for broader access to experimental medicines to deal with the new and increasing threat of Ebola, now a WHO public health emergency.
We would like to suggest one approach to be considered. The urgency of this threat presents us with a new opportunity: To conduct such research openly, in order to accelerate the science through the contributions of the broader community.
A group of scientists, entrepreneurs, and funding agencies recently met to discuss the possibility of open source pharma and a vision of Medicine for All, in which we argue that the the fastest and most efficient way to make progress against emerging health threats is in an open arena, permitting rapid collaboration between industry, academia, and the public.
The signatories of this letter, which include some attendees from that meeting, would like to suggest that in your consultations with colleagues and funding agencies in the coming days, that you ask whether we could conduct the research openly, so that all data and ideas are shared and anyone may participate. We have already received specific and multiple expressions of interest from highly qualified groups for working in this way to:
Critically assess current trial medicines through crowdsourcing and to clarify the public state of knowledge. There is scant information on this outside pay-walled scientific literature. This assessment would be achieved through the creation of an open knowledge forum, in which all data related to the current drug candidates and relevant research will be collated and shared, such that scientists worldwide have access to this knowledge and can build upon this resource.
Investigate new or re-purposed therapies experimentally via novel coordinated screens of diverse molecules in existing collections to be performed in the short term with complete release of data to the community. Working in the public domain will improve coordination and reduce duplication of resources.
Design new clinical study protocols using existing online resources donated for this purpose, again using the knowledge of diverse professionals collaborating to solve a problem quickly.
Engage broader communities by leveraging crowdfunding to increase awareness and provide matching funds for research projects.
Once open research on Ebola gains support at the highest levels, we are sure that further initiatives and resources would be forthcoming, in addition to the ones above. Also, opening the process of knowledge generation to the public will encourage companies owning certain IP rights to contribute to the discussion.
What is needed to make this happen is simply strong political will followed by modest, short-term funding--funding that might be more likely to arise from other organizations following WHO support. Most important in the short term, to galvanise the already-interested community, is for WHO to state that a complementary open approach has intrinsic merit. We believe the approach is aligned with the pilot projects and the Consultative Expert Working Group.
Since the public at large have been asking the question “Why are there no drugs for Ebola?” and are perhaps unaware that open research into new medical treatments is being supported by a diverse set of people, we are posting this letter openly to raise awareness.
Sherwood Neiss, who attended the Open Source Pharma Conference at the Bellagio Center and is a principal at Crowdfund Capital Advisors, wrote a blog post explaining how crowdfunding could benefit the open source pharma movement.
"Crowdfunding isn't a panacea. Taking into account all forms of crowdfunding, less than 35% of initiatives are successful with their fundraising. However, it does increase the opportunity for individuals' to access capital where the traditional financiers have either shown no interest in funding or have stepped out of the equation. Depending on the type of crowdfunding it requires that certain technology, social, cultural, regulatory pieces are in place."
Neiss goes on to suggest several tactics for setting up a successful crowdfunding initiative, including:
"Seek matching funds or create an X-Prize type of reward to further encourage collaboration and move development through the pipeline. Reward people along the way and promote multiple smaller campaigns rather than one large one. This will continually engage backers along the development lifecycle."
"To meet the changing needs of the biomedical workforce, NIH is piloting the concept of awarding longer grants that provide more stable support for investigators at all career stages. It is our hope that with more sustained support, investigators will have more freedom to innovate and explore new lines of inquiry. The NIH Pioneer Award, supported by the Common Fund, represents a compelling example of such an approach. Pioneer Awards support individual scientists of exceptional creativity, who propose pioneering—and possibly transformative approaches—to major research challenges. This award allows for $500,000 annually in direct costs for five years, and, in a recent evaluation, was shown to facilitate a high level of innovation and productivity.
Moving forward, several NIH Institutes and Centers (IC) will be developing new funding opportunities to offer more sustained support to investigators’ research programs. These longer term awards will not follow a one-size-fits-all approach; leaders of each NIH IC will decide if they wish to embark on these awards based on the balance of their portfolios and their strategic planning needs. In addition, each IC will decide the appropriate size and duration of their awards. While applications for these awards will not require specific aims in the traditional R01 format, investigators will describe their research plans and will demonstrate how they will leverage and translate their prior accomplishments into approaches that will shape their future research."
MyScienceWork.com: "Sylvain Allano, head of science and future technologies at PSA Peugeot Citroën was present that morning to introduce the strategy of open innovation of the group. “The car industry is facing huge challenges that will change it completely in some years,”he said. The evolution of urban mobility, protection of the environment, and users’ new expectations are all part of these challenges to come. “The ones courting us nowadays are people from Silicon Valley, like Google and Facebook.” Why are these huge internet firms interested in PSA? Tomorrow’s connected vehicles will become clusters of information, sucking up data as they carry people, energy and information.
To imagine that vehicle of the future, PSA established a policy of open innovation 5 years ago, which was made concrete with open innovation centers, in France and abroad, in partnership with universities. The economic crisis and increasingly complex technological challenges are so many reasons to integrate actors from outside the company into the innovation effort: “Before, we needed to look for specific knowledge to solve a technical issue. Today, the game is not so much about finding solutions as asking the right questions. That is why we are looking elsewhere, opening a very broad scientific field.” Open innovation thus needs a collaborative spirit to exist, leaving the doors open to creativity, from the academic world, scientific institutions and independents."
Google and Novartis to create Glucose-Monitoring Contacts for Diabetics
TechCrunch: "Google has announced a new licensing deal with big pharmacy corporation Novartis to bring its “smart” contact lenses, with the power to help diabetics track their blood glucose level, to patients. The commercial launch of the lenses will take place through Novartis’ Alcon eyecare division and Novartis CEO Joe Jimenez told the Financial Times that a timeline for commercialization that took longer than five years would be a 'disappoint[ment].' "
"While most major tech companies are entering into the health market in some way, this partnership by Google represents a completely different approach to the potential opportunity. Novartis’ CEO told the FT that the market could be worth anywhere between $10 billion and $50 billion in the next ten years, and Google’s assistance in building not only blood glucose sensing eyewear (which addresses the ballooning potential client base of diabetics in the U.S.), but also the corrective eyewear market could help it capture a sizable piece of that pie."
What is Open Access? A Handy Guide from the Open Society Foundations
8/10/2014 "Open access is an alternative publishing and distribution model that makes scholarly research literature—most of which is already funded by taxpayers around the world—freely available to the public online, without restrictions.
Harnessing the power of the internet, open access brings the results of academic research to unprecedented numbers of scientists, university professors, medical researchers, patients, inventors, students, and others."
Report: Establishing Incentives and Changing Cultures to Support Data Access
Expert Advisory Group report on Data Access. EAGDA was established by the MRC, ESRC, Cancer Research UK and the Wellcome Trust in 2012 to provide strategic advice on emerging scientific, ethical and legal issues in relation to data access for cohort and longitudinal studies.
"EAGDA wished to understand the factors that help and hinder individual researchers in making their data (both published and unpublished) available to other researchers, and to examine the potential need for new types of incentives to enable data access and sharing. This is a critical challenge in achieving the shared policy commitment of the four EAGDA funders to maximise the benefit derived from data outputs and the considerable investment they have made over recent years in supporting data sharing.
In addition to a review of previous reports and other initiatives in this area, the work involved in-depth interviews with key stakeholders; two focus group discussions; and a web survey to which 35 responses were received from a broad range of researchers and data managers."
EAGDA findings from the Wellcome Trust website:
making data accessible to others can carry a significant cost to researchers (both in terms of financial resource and the time it requires) and there are constraints in terms of protecting the privacy and confidentiality of research participants;
while funders have done much valuable work to encourage data access and have made significant investments to support key data resources (such as the UK Data Service for the social sciences), the data management and sharing plans they request of researchers are often not reviewed nor resourced adequately, and the delivery of these plans neither routinely monitored nor enforced;
there is typically very little, if any, formal recognition for data outputs in key assessment processes – including in funding decisions, academic promotion, and in the UK Research Excellence Framework;
data managers have an increasingly vital role as members of research teams, but are often afforded a low status and few career progression opportunities;
working in data intensive research areas can create potential challenges for early career researchers in developing careers in these fields;
the infrastructures needed to support researchers in data management and sharing, and to ensure the long-term preservation and curation of data, are often lacking (both at an institutional and a community level).
"The CEO of German pharmaceutical company Bayer, Marijn Dekkers, was reported as saying that the company didn’t develop a cancer drug for the Indian market, but rather 'for Western patients who can afford it.' The comment summed up the attitude of the pharmaceutical companies towards the poor and succinctly described what is wrong with today’s research and development (R&D) system.
In a similar vein, last month British/Swedish pharma company AstraZeneca announced it was pulling out of all early-stage R&D for malaria, tuberculosis (TB) and neglected tropical diseases – all diseases of the developing world. Instead, the company stated they will focus efforts on drugs for cancer, diabetes and high blood pressure, all diseases that affect rich countries, with potentially plenty of people to pay the high prices on new drugs."