November 5, 2015

Prize4Life, Sage Bionetworks and DREAM Announce Winners of ALS Stratification Challenge

Prize4Life, Sage Bionetworks and DREAM Announce Winners of ALS Stratification Challenge

SEATTLE–(BUSINESS WIRE)–Today, Prize4Life, Sage Bionetworks and the DREAM community announced the winners of the DREAM ALS Stratification Prize4Life Challenge, a global data analysis competition to computationally identify different subgroups of ALS patients. In ALS, there is currently no way to determine whether someone living with the disease will survive the average of 3-5 years after diagnosis or live with ALS for decades. The goal of the Challenge is to find answers that may inform clinical practice, new trial designs, and ultimately personalized approaches to the discovery and development of new ALS medicines.

“We want to accelerate ALS drug discovery and believe that critical breakthroughs will come from new ways of thinking, collaborating, and sourcing ideas,” said Shay Rishoni, CEO at Prize4Life. “I’m excited by the results of the ALS Stratification Challenge and congratulate the winners who significantly outperformed the bar.”

Participants were free to compete as individuals or on teams and could work on as many as 4 different sub-challenges.

In Sub-challenge 1, the winner, Team UglyDuckling is from the Department of Computer Science and Information Engineering at the National Cheng Kung University, Taiwan. The team comprises Dr. Wen-Chieh Fang, Huan-Jui Chang, Chen Yang, Prof. Hsih-Te Yang, and Prof. Jung-Hsien Chiang. Dr. Wen-Chieh Fang, a post doc at Prof. Jung-Hsien Chiang’s lab, is the leader. This is Team UglyDuckling’s second DREAM Challenge participation and first win.

In Sub-challenge 2 and Sub-challenge 4, the winner, Team Guanlab_Umich, is from the Department for Computational Medicine and Bioinformatics at the University of Michigan, USA. Prof. Yuanfang Guan is a faculty member, and having won seven DREAM challenges (three in 2015 alone), is an emerging leader in machine learning.

In Sub-challenge 3, the winner, Team Jinfeng_Xiaofrom is from the Center for Biophysics and Quantitative Biology at the University of Illinois at Urbana-Champaign, USA. Jinfeng Xiao is a second year PhD student and repeat Challenge winner. This is Xiao’s first individual win.

“We thank the participants of the DREAM ALS Stratification Prize4Life Challenge,” said Stephen Friend, President, Co-Founder, and Director of Sage Bionetworks. “ALS is a variable disease and the speed of progression is different from person to person. Through our community-based approach, our participants were able to mine the largest ALS clinical trials and registry databases and produce promising computational models to find patterns and identify subgroups of ALS patients.”

The winning teams will share a $28,000 cash prize, be invited to present their winning models at the DREAM Conference on Regulatory and Systems Genomics meeting on November 16-18, and have the opportunity to co-author a Challenge overview paper for which the journal partner is Nature Biotechnology.

The ALS Stratification Challenge is a unique competition defined by the diversity of its participants and resourceful sponsorship. The Challenge featured over 70 entrants from 30 teams across 15 countries and was sponsored by leading research-based biopharmaceutical companies Biogen and Eli Lilly and Company. IBM also supported the Challenge with computational resources. The cash prizes were raised by running a crowdfunding campaign called “Fund the Prize” that ran in winter 2014-2015 and succeeded in raising $28,000 for prizes.

“The Challenge winners and participant community will help deepen our understanding of ALS, a devastating and complicated disease,” says Gustavo Stolovitzky, DREAM founder from IBM Research and the Icahn School of Medicine at Mount Sinai. “Crowdsourcing a powerful dataset and having participants submit their source code to predict patient prognosis has advanced the Sage Bionetworks’ and DREAM’s mission of fostering open and collaborative science while addressing the problem of stratification in ALS.”

Prize4Life provided the largest open-access ALS clinical trials database in the world (PRO-ACT:, developed with NEALS and MGH, which served as the basis for the Challenge along with national ALS registries from Ireland and Italy. Sage Bionetworks and DREAM created a cloud-based Challenge platform called Synapse ( where participants accessed the data and shared ideas.

Challenge participants performed work on cloud-based computational resources donated by IBM. An exciting technical feature of this Challenge was that participants submitted their open source algorithms in a portable framework to the IBM cloud for scoring. At the end of the Challenge all of the submitted analysis methods will be available on Sage’s Synapse platform as a Challenge library and community resource that can serve as the basis of ongoing research.

“We are committed to finding answers to ALS and dedicated to supporting new approaches like crowdsourcing and big data analysis that have the potential to provide powerful insights into the disease,” said Donald R. Johns, M.D., Vice President, ALS Development at Biogen. “We are grateful to the participants of the DREAM ALS Stratification Prize4Life Challenge who have created new knowledge and an important repository for research that may bring improved trials, new treatments, and hope to ALS patients.

Press Release on BusinessWire