Saez-Rodriguez Group

Systems Biomedicine

Crowdsourcing computational biomedicine – DREAM challenges

The main challenges in systems biomedicine are very complex, and can  not be fully solved by a research group alone. Therefore, we are involved in DREAM challenges, a community effort to advance our understanding of fundamental problems in systems biology and translational medicine.

We ask these questions to the whole scientific community as a collaborative competition whereby, together with specific data providers, we pose questions and provide the data necessary to address them, such as what is the best drug to treat a tumour type. Anyone can participate, and DREAM provides an unbiased, rigorous assessment of a team’s solution. When the challenge is closed, participants’  solutions are analyse to learn what methods did well (Saez-Rodriguez et al. 2016).

DREAM was founded in 2006 by Gustavo Stolovitzky ( IBM Research) and Andrea Califano (Columbia University). Julio is DREAM co-Director for Computational Systems Biology Challenges, and we have been involved in a number of challenges, ranging from infererence of signaling networks from phosphoproteomic data(Hill et al. 2016; Prill et al. 2011; Meyer et al. 2014) to predicting the efficacy or side effects of drugs alone or in combination (Costello et al. 2014; Eduati et al. 2015; Menden et al. 2017). The latest challenge we co-organized, with the NCI-CPTAC Consortium, tried to shed light on whether it is possible to predict (phospho)proteomic data from genomic and transcriptomic data.

See below a video about a recent DREAM challenge (Eduati et al., Nat Biotechnol. 2015) on toxicogenomics:

Selected references: