Adam Margolin launched the Sage Bionetworks DREAM Breast Cancer Prognosis Challenge with an introductory webinar on July 17. We were so excited to see this Challenge launch with nearly 200 participants coming from 20 different countries! As part of the Sage Bionetworks’ mission, this initiative seeks to catalyze more open and collaborative approaches to advancing scientific research. The specific goal of this Challenge is to to assess the accuracy of computational models to predict breast cancer survival using patient clinical and genomic data. Adam’s webinar presentation includes background on the Challenge and a live demonstration of how to access the Challenge data, build prognostic models, and display performance results on a real time leaderboard. See the challenge website for further information and you can track participant progress on the real time leader board.
George Church from Harvard University provided a great perspective on the accomplishments and future of genomic medicine in his keynote address at the 3rd Commons Congress entitled, “Personal Genome Project and Beyond”. Copies of the presentation can also be downloaded from the Congress agenda page.
David Haussler from the University of California, Santa Cruz gave an exciting update on bioinformatics advances in his keynote address entitled, “Personal Cancer Genomics” at the 3rd Commons Congress. Copies of the presentation can be downloaded from the Congress agenda page.
Larry Lessig from Harvard University presented an entertaining and engaging keynote talk at the 3rd Commons Congress entitled, “Ingredients for innovation” that highlighted the systematic problems with intellectual property protection laws in particular and the abusive way in which money influences government in general.
Adrien Treuille from Carnegie Mellon University is the recipient of the first Leland Hartwell Award for Innovation in Open Networks. He presented a keynote presentation at the 3rd Commons Congress on the 10 lessons he has learned using game playing to engage large numbers of citizen and professional scientists in cutting edge research. Copies of the presentation can be downloaded from the Congress agenda page.
Rick Klausner from The Column Group gave a wonderfully irreverent and provocative keynote talk at the 3rd Commons Congress entitled, “New Models for Open Innovation and the Crisis of Translation”. Copies of the presentation can be downloaded from the Congress agenda page.
Jamie Heywood from PatientsLikeMe gave a keynote address at the 3rd Commons Congress entitled, “Discovery 2.0 – I don’t know how to solve this problem but we might”. Copies of the presentation can be downloaded from the Congress agenda page.
Mike Kellen examined the essential framework and features of the Sage Bionetworks collaborative IT platform “Synapse” to align experts and enthusiasts for the next directions to be taken as well as describing the next functionalities to be built. Highlights included: Synapse as way to publicly host data for broad re-use, real time analysis done in Synapse and used to support a journal article. Commentators included: Robert Gentleman, Sean Hill, Jason Johnson, Dietrich Stephan, Magali Haas
Lara Mangravite & John Wilbanks discussed how patients have traditionally been sidelined in medical research and need to be enabled to become activated partners. They highlighted the current consent and privacy barriers along with multiple pilot projects to allow patients and scientists to work together in sharing their data. Such opportunities require new governance models for sharing data, tools and disease models. Commentators include: Kelly Edwards, Peter Kapitein, Jane Kaye, Sharon Terry
Portable Legal Consent
Jonathan Derry & Adam Margolin described highlights from the past year of success in building “networked” models of diseases to stimulate discussions on potential projects for next year.
IOM Precision Medicine Report
Cell Line Encyclopedia
Adam Margolin detailed existing and emerging collaborative ‘challenge projects’ being developed by Sage Bionetworks, METABRIC, Norway Radium Hospital, IBM, and Google. He noted that citizens need to be directly involved in networked approaches to solving biological problems such as finding the causes of disease, discovering new therapies and identifying who should be given which drug. Citizen science requires new modes of communication and collaboration that go beyond traditional paradigms. It also requires nontraditional funding strategies. Commentators included: Regis Kelly, Dan Gallahan, Henry Chesbrough, Ilya Kupershmidt
Vicki Seyfert-Margolis from the Food & Drug Administration provided an update on the role of the FDA to build large public disease biology-based clinical and genomic datasets. She presented approaches to extend beyond the 2011 pilot project of accessing the “clinical trial comparator arms” of industry clinical trials (CTCAP) and examined new joint opportunities in regulatory sciences.
Stephen Friend introduced a panel discussion at the 3rd Commons Congress focused on cutting edge open solutions to traditionally-siloed drug discovery models: Among those to be highlighted will be the COMPASS trials, Cinderella Therapeutics, Structural Genomics Consortium, The Discovery Network, Arch2POCM, and the Cancer Commons. This crash course on Public-Private Partnerships and other ways to de-risk therapeutic targets included comments from Barbara Mittleman, Huib Vriesendorp, Thea Norman, Kathy Giusti, Rob Snelders, and Marty Tenenbaum
SCG & Arch2POCM
Arch2POCM in Science Trans Med
Structural Genetics Consortium
Chitra Krishnan, Michael Simpson & Tyler Jenks announced the creation of a new IT platform to accelerate research. Current reward structures in the medical industrial complex marginalize citizens and patients and new ways to bring together citizens, patients, funders, and scientists are needed. BRIDGE is a new approach pioneered by Ashoka/Sage Bionetworks to facilitate non-traditional support of innovative medical research. Commentators included: Jamie Heywood, Bas Bloem, Bev Mayhew, Eva Guinan, Vessela Kristensen, Laura Van’t Veer
BRIDGE Project Description
Stephen Friend & John Wilbanks led the final session of the 3rd Commons Congress where participants focused on the projects and components and to ask what opportunities are worth resourcing to promote sustainability.
Following the theme “Building Better Models of Diseases Together”, the Congress was organized to discuss redefining tools needed, how we share and our roles as researchers citizens and funders. In the spirit of the “Continental Congress in Philadelphia” the workshop groups shaped the strategy and directions of project execution, mechanisms for outreach, and policy change needed for a systems change that will enable networked approaches to building better models of diseases. The goal of the final session was to build a roadmap of how to collectively overcome the weaknesses and gaps in our current and adjacent complementary efforts and to anticipate what is needed to build a Commons where activated citizens can jointly contribute as patients, as funders and as researchers – a democratization of medicine.
Bas Bloem, Parkinson Net
I describe how research should transform from its traditional approach – where scientists dictate the research agenda and possess the data – to a new world of ‘Research 2.0’ or ‘participatory research’ where patients define what they feel is clinically most relevant, where patients own their own data, and where scientists collaborate in order to best answer the patients’ questions
Phil Bourne, PloS ONE, SDSC
Much of science is closed and hierarchical. Occasionally something happens that really drives home how limiting that can be. Such is the story of Meredith. Citizen scientists typically contribute data to be interpreted by the experts. Garage scientists run their own experiments which occasionally make it big (think Apple). Enter “digger scientists” who turn apparent digital trash into treasure.
Geoffrey Clapp, Rock Health
Telemedicine has been slow to be adopted at “internet scale,” in many cases due to legislative hurdles and medical licensure. Several legislative changes are currently in progress that will not only make telemedicine more attractive, providing the rocket fuel to the telemedicine, opening up many new revenue sources for doctors and startups in the health/tech space, and better, more cost effective care for patients.
Krishanu Saha, MIT
Embryonic-like stem cells can now be induced from routine blood samples from nearly any person. These cells grown in Petri dishes may used to model disease in the individual and thus are called a “disease-in-a-dish”. But has the individual been lost in this categorization? I will describe how various individuals – researchers, doctors, patients, citizens, donor-participants, and future consumers of personalized medicine – emerge in the research architectures constituting this field.
Markus Warmuth, H3 Biomedicine
Lack of access to powerful and expansive experimental platforms and infrastructure essential for oncology target validation and continued hypothesis validation from the drug discovery process and up until clinical trial significantly hinders small or medium size biotechnology companies, research or academic institution’s capability to advance oncology drug discovery. We propose the idea of Translational Pre-Clinical Oncology (TransPOC) Consortium, which seeks to establish translational infrastructure including experimental platforms and novel technologies for information/data sharing, mining capability and dynamic research/publishing through shared resources.
Aled Edwards, Structural Genetics Consortium
Bibliometric analysis of research activity reveals that most research in academic and industry is focused on 10% of human proteins, despite genetic evidence highlighting disease-related proteins in the “unstudied” 90%. Even if we succeed in describing disease networks and in uncovering new targets, will the peer-review system allow us to study them? Is our cherished peer-review system an anachronism?
Larry Hunter, University of Colorado, Denver
Traditional analytics need defined outcomes to build predictive models. Theories of reinforcement learning, particularly temporal difference learning, have the potential to help patients use high-frequency phenotyping to manage their disease, maximizing diverse sets of positive events while minimizing negative ones.
Kathy Giusti, Multiple Myeloma Research Foundation
The emerging technological and digital revolution has had a significant impact on patients’ ability to advance progress toward personalized treatments and cures. I will provide examples of how the Multiple Myeloma Research Foundation is developing and implementing models and programs to leverage these opportunities
Francois Grey, CERN
Like medieval guilds experts require long training and are paid well. Recent crowd sourcing methods have shown an alternative. Is there an opportunity for crowdsourcing parts of the expertise needed to treat patients?
Heather Joseph, SPARC
There is a growing movement to change the lack of taxpayer’s access to scientific research by supporting policies that require all published results to be made available to the public, free, online as soon as possible. This talk will outline actions you can take to get involved and make taxpayer-funded science openly accessible.
Tom Insel, NIH
Translational science relevant to drug discovery has hit a wall. When we shift our focus from “what science we do” to “how we do science”, several solutions emerge. Perhaps most critical is the recognition that partnerships built around pre-competitive sharing of compounds, data, and resources can accelerate a key step in translation — target validation. NIH via its new National Center for Advancing Translational Science (NCATS) can catalyze translation by changing “how we do science”
Misha Angrist, Duke University
I discuss two experiences I have had as a member of the Duke IRB, both of which illustrate how far we are from changing the culture of academic research from one based on paternalism, genetic exceptionalism and legal liability to one based on autonomy and partnership. It’s hard to be optimistic sometimes.
Iya Khalil, GNS Healthcare
A near fatal bilateral pulmonary embolism has given me a unique perspective on the state of healthcare and what the biomedical community needs to achieve beyond sequencing our genomes to ultimately help the end consumer in this system – the patient
Joseph Hellerstein, Google
Scientific discovery is in transition from a focus on data collection to an emphasis on analysis and prediction using large scale computation. These computations can be done with unused cycles in commercial Clouds if there is appropriate software support. This talk describes the Google Exacycle system that routinely provides more than 100K cores for scientists doing computation based discovery.
John Schrom, Rock Health
The average 24-year-old will spend more time on Facebook in the next week than they will spend with a physician in the next 20 years. I will provide examples of how our health is expressed through social media, how we can use that to improve lives and fight disease, and why this is the start of a transformational change in public health and medicine.
Francois Taddei, INSERM Paris
Learning through research is a powerful educational tool that is often limited to students in research universities. Can open science, open technology, and open education converge as a force to democratize and reinvent how learning through research is done? I provide an example of ways forward that I would like to co-design with the sage community.
Susan Love, Susan Love Research Foundation
All efforts to revolutionize our approach to medical research are based on the same silos of researchers and physicians on one side and patients on the other. The term patient refers to what is usually experienced as a one down position. This is an artificial dichotomy since we are all patients or soon to be patients. This should be a public effort with all of us bringing our expertise to the table to work together to eradicate disease. Sometimes we may be wearing a doctor hat but we also may have a researcher and a patient hat…they really are not different groups. We are all now or soon to be patients!
Dr. Kelly Edwards from the University of Washington explores the foundations of sharing for research scientists and the critical importance of trust. She uses her expertise in bioethics to look at the contractual and implied components and motivations behind effective, altruistic collaborative research.
Dr Erich Huang, the host of Conversations on Network Building in the Big Data Deluge, discusses tools he has created to enable a community of disease pathway modelers looking for genetic signatures of cancer. He describes shotgun stochastic search methods and bayesian analysis in the context of the Synapse compute platform for integrating multiple known genetic pathways into informative disease signatures.
Dr. Jerry Radich from the Fred Hutchinson Cancer Research Center looks at the progress being made on CARDINAL, a project to distinguish patients who will respond to conventional adult AML therapy from those who will need more agressive strategies such as bone marrow transplantation. Dr. Radich relates his experience with multilevel barriers such as regulatory requirements, funding problems, academic ecosystem pressures that need to be addressed for effective collaborative research.