What is the ALS Prediction Prize and why are the solutions so important?
The DREAM-Phil Bowen ALS Prediction Prize4Life Challenge aims to confront a basic puzzling question in ALS: most patients are like Lou Gehrig, with a rapidly progressing disease course. Some patients, however, turn out to be more like Stephen Hawking, where the disease progression is delayed. What separates the Lou Gehrigs from the Stephen Hawkings? Within the ALS patient population there is enormous variability with some people living for many years or even decades, while others die much sooner. This makes it extremely difficult to develop new and effective treatments for this as-yet incurable disease. Solving this mystery is important, for patients and their families and for those planning clinical trials of potential new treatments. The winners of the ALS Prediction Prize may hold the key.
On average, people diagnosed with ALS, a fatal disease, live about 1,000 days (around three years). Therefore, it is extremely difficult to develop new and effective treatments for this as-yet incurable disease. Currently, there is only one FDA approved drug available for ALS patients. The drug does not improve quality of life and only extends life by about three months.
How were the winners of the ALS Prediction Prize able to determine their solutions?
The winning solvers of the ALS Prediction Prize have developed algorithms that predict a given patient's disease status within a year's time based on three months of data. This solution is important because it could impact how clinical trials for ALS therapies are designed and conducted, fostering faster breakthroughs in effective treatments for the disease.
Registered solvers were provided a small subset of data from the PRO-ACT database, the largest database of clinical data from ALS patients ever created. The fully anonymized data includes patient demographics, medical and family history data, functional measures, vital signs and lab results. The full set of data from over 8,500 ALS patients will be globally available for research purposes beginning December 5, 2012.
The Prediction Prize is a powerful example of how "Big Data" can lead to improved advances in medicine. Anyone with quantitative abilities, be they an engineer or atmospheric chemist, can help in the fight against ALS.
Why did we use a crowdsourcing approach for the ALS Prediction Prize?
Currently, ALS trials must include large numbers of patients to account for the enormous variance in the course of the disease progression within the ALS patient population, making these trials costly, slow and difficult to interpret. By making clinical trial data available to a global community of data scientists, researchers, and computer mavens, we are speeding up the process while driving down the costs of discovery, which is good news for both the scientific and patient communities we serve.
Prize4Life's mission is to accelerate the development of treatments for ALS using a prize-for-breakthrough model. The ALS Prediction Prize follows the success of the $1 million ALS Biomarker Prize4Life awarded earlier this year; both are examples of our organization successfully using crowdsourcing to encourage scientific and medical breakthroughs.
Who are the winners?
Two teams have secured first place in the ALS Prediction Prize: a duo from Stanford University, postdoctoral candidate in mathematics and statistics Lester Mackey, and recent JD and Master's Degree recipient Lilly Fang; and the team of Liuxia Wang, Principal Scientist, and her colleague Guang Li, Quantitative Modeler at Washington, DC-based scientific marketing company, Sentrana. Each team will receive $20,000 for generating the top-performing solutions to predict disease progression in ALS patients.
In addition, Torsten Hothorn, a distinguished statistics professor from Germany, was awarded a second-place prize of $10,000 for his unique solution which included an alternative approach to assessing disease progression to that specified in the challenge criteria. The ALS Prediction Prize judging panel found that using this alternative method could potentially yield highly impactful results, so the organization created a second-place prize to recognize Hothorn's innovative thinking.
Why did Prize4Life choose three winners?
We saw an overwhelming response to the competition, and the efforts from data scientists worldwide resulted in at least three viable (and potentially complementary) solutions. For ALS patients and their families, this is a huge win and a very promising step toward effective treatments for ALS.
Among the many proposed solutions submitted over the Innocentive platform, the solutions offered by the Wang/Li and Mackey/Fang teams scored virtually identically, even though both the statistical methods and the parameters chosen by each team were different; in addition, the solution offered by Hothorn scored extremely closely to that of the other two teams.
Over 1000 individuals and teams registered to participate in the challenge, 25 of them submitted complete algorithms. Given the quality of the results submitted, our judge's panel realized it was impossible to award just one prize as we had originally planned. With the help of a generous donor deeply committed to helping find a cure for ALS, we decided to double the prize amount we had initially allocated for the winning solutions.
What do these solutions mean to the search for treatments and a cure for ALS?
The solutions to the ALS Prediction Prize bring us one step closer to effective treatments for ALS. Currently, it is impossible to know upon being diagnosed with ALS how long a patient will live. New prediction tools, such as those developed by the winners of the ALS Prediction Prize, give scientists and medical experts another weapon in their arsenal to use in the fight against ALS.
Prize4Life is a non-profit organization and our mission is to accelerate the development of treatments for ALS using a prize-for-breakthrough model. Prize4life awarded $1M to the winner/solver of the ALS Biomarker Prize4Life Challenge and is now awarding $50,000 to the winners/solvers/ of the ALS Prediction Prize; both are examples of the organization successfully using crowdsourcing to encourage scientific and medical breakthroughs.
Who did Prize4Life partner with?
Prize4Life partnered with three organizations for the ALS Prediction Prize:
- InnoCentive - the world's largest prize platform, which has an international solver network numbering in the millions
- The DREAM Project - an international organization of computational biologists dedicated to open access data and challenges addressing important scientific problems (organizers of RECOMB conference). http://www.the-dream-project.org/
- The family of Phil Bowen - Mr. Bowen's son Peter, who was also involved in the start-up phase of creating the PRO-ACT database, raised a large amount of funding and played an integral role in garnering attention and awareness for the prize in honor of his father, who died of ALS http://fundraise.prize4life.org/e/pbowen
Who are the ALS Prediction Prize Judges?
Merit Cudkowicz M.D. MSc. is the Julianne Dorn Professor of Neurology at Massachusetts General Hospital, at Harvard Medical School. Dr. Cudkowicz completed medical training at the Health Science and Technology program of Harvard Medical School, and she was a resident in Neurology at MGH. She obtained a Master's degree in Clinical Epidemiology from the Harvard School of Public Health. Dr. Cudkowicz's research and clinical activities are dedicated to the study and treatment of patients with neurodegenerative disorders, in particular amyotrophic lateral sclerosis (ALS). Dr. Cudkowicz directs the MGH ALS clinic and the Neurology Clinical Trials Unit . She is one of the founders and co-directors of the Northeast ALS Consortium (NEALS), a group of 92 clinical sites in the United States and Canada dedicated to performing collaborative academic led clinical trials in ALS. In conjunction with the NEALS consortium, she planned and completed 7 multi-center clinical trials in ALS and is currently leading three new trials in ALS. Dr. Cudkowicz received the American Academy of Neurology 2009 Sheila Essay ALS award. She is actively mentoring young neurologists in clinical investigation. Dr. Cudkowicz is on the Research Council of the American Acadenmy of Neurology and the medical advisory boards for the Muscular Dystrophy Association.
Orla Hardiman BSc MB BCh BAO MD FRCPI FAAN is an consultant neurologist. She is a HRB Clinician Scientist, Clinical Professor of Neurology at the University of Dublin and a Consultant Neurologist at the National Neuroscience Center of Ireland at Beaumont Hospital, Dublin. Hardiman has become a prominent advocate for neurological patients in Ireland, and for patients within the Irish health system generally. She is co-Founder of the Neurological Alliance of Ireland, and Doctors Alliance for Better Public Healthcare. Hardiman is current Dean of the Irish Institute of Clinical Neuroscience. In the past, she established the bi-annual Diaspora Meeting, a forum for Irish neurologists based overseas to present and discuss their research findings with neurologists working in Ireland
Robert Küffner habilitated in informatics in 2010 and is currently a group leader for computer science and bioinformatics at the Ludwig-Maximilians Universität München. He received his PhD in molecular biology in 1998 at the Heinrich-Heine Universität in Düsseldorf, Germany. His main interests include the analysis of biological networks via Petri Nets as well as the areas of text mining, expression analysis, gene regulation, and systems biology. Recently, Dr. Küffner's team was recognized as the best performer in two international community-wide computational challenges where comprehensive blinded assessments of network inference approaches have been conducted.
Raquel Norel is part of the Functional Genomics and System Biology Group in IBM Research where she uses math and computing to bring insight to complex biological problems. Recently she has been working on collaboration-by-competition projects; since 2010 she has contributed to the DREAM project as an organizer and scorer. She contributes regularly to Faculty of 1000. Raquel holds a PhD in Computer Science from Tel Aviv University, where she wrote her thesis on "Algorithms for Protein Docking" while co-authoring 10 papers on the subject. She also has an MSc in Computer Science from Weizmann Institute of Science, with a thesis on "A Model for the Adjustment of the Mitotic Clock by Cyclin and PMF levels"; this worked was published in Science. She earned a BSc in Engineering Sciences from Universidad de Chile.
Gustavo Stolovitzky received his M.Sc. in Physics (with honors), from the University of Buenos Aires (1987) and his Ph.D. in Mechanical Engineering from Yale University (1994), which awarded him the Henry Prentiss Becton Prize for Excellence in Engineering and Applied Sciences. In 1998 he joined the IBM Computational Biology center at IBM Research where he is the manager of the IBM Functional Genomics & System Biology Group.
Gustavo has had an active role in organizing the systems biology community. He founded and leads the DREAM project, an international effort that has nucleated thousands of participants to assess the performance of systems biology methods. He also co-organizes the RECOMB Systems and Regulatory Genomics and DREAM challenge conferences, which have attracted around 500 attendees every year for the last 4 years. He has co-authored more than 100 scientific publications, has 9 issued patents, and has edited 2 books. His work has been highlighted in The New York Times, The Economist, Technology Review, and Scientific American (where his DNA transistor project was chosen as one of the 10 world changing ideas of 2010) among other media. Gustavo is a member the PLoS ONE and the OMICS editorial board, and has been elected fellow of the NY Academy of Sciences, fellow of the American Physical Society, fellow of the American Association for the Advancement of Science, and fellow of the World Technology Network. He holds a position as an adjunct Associate Professor at Columbia University.
His most recent scientific interests are in the field of high-throughput biological-data analysis, reverse engineering biological circuits, the mathematical modeling of biological processes, and next generation technologies for DNA sequencing.