NEW YORK (GenomeWeb News) – A group of federal, academic, and industry partners in the US have launched a new challenge to spur the development of new computational tools for predicting patient responses to certain drugs and chemicals based on their genomes, Sage Bionetworks said today.
Aiming to create tools to address the massive backlog of chemicals that have not been tested for toxicity, participants in the Toxicogenetics Challenge will use crowd-sourcing approaches to work with newly released datasets of cytotoxicity testing of 179 pharmaceutical and environmental chemicals.
Along with Sage, the competition was launched with partners including the National Institute of Environmental Health Sciences, the National Center for Advancing Translational Sciences, and the University of North Carolina at Chapel Hill, and the Dialogue for Reverse Engineering Assessments and Methods, or DREAM, a public-private partnership that seeks to initiate systems biology competitions.
The partners said that the new cytotoxicity data, which is being made available through the 1,000 Genomes Project, provides unique datasets that researchers could use to predict toxic responses to compounds in a genetically diverse human population, particularly when the data is matched with other available genomic data.
"The long-term strategic value of accurate predictive models will be invaluable for both protection of human health and the environment, and support of innovations in the chemical industry," Ivan Rusyn, a professor at UNC's Gillings School of Global Public Health, said in a statement.
Participants in the challenge will be asked to solve one or both of two sub-challenges: to use the toxicogenomic data to develop a model to predict individual response to compounds based on genomic information and to develop a model to predict how a particular population will respond to certain types of chemicals.
Sage and DREAM's organizers will release tools and incentives during the three-month competition, and they will pose sub-challenges and maintain a leader board with "scores" of the predictions that the competitors submit. The organizers also seek to foster collaboration by creating rewards that will encourage the challengers to submit some code for their models that will enable others to build hybrid models.
The winners of the Toxicogenomics Challenge will be awarded at a conference this September in Toronto, and travel grants will be awarded to the top performing teams. The funding for travel is being provided by DREAM partners at Columbia University and IBM Research.