The organizers of the annual Dialogue for Reverse Engineering Assessment and Methods (DREAM) have launched three new modeling challenges focused on Alzheimers, cancer, and rheumatoid arthritis.
The first challenge provides participants with cognitive, imaging, biological, and whole-genome sequencing data from the Alzheimer’s Disease Neuroimaging Initiative, and asks them to build two models, one which "predicts change over time in AD cognitive scores using all available test and adjacent data" and another that "predicts discordance between biomarkers suggestive of amyloid perturbations and lack of cognitive impairment."
For the second challenge, participants will have nine terabytes of raw human sequence data from prostate and pancreas tumor-normal tissue pairs, and will work on models that accurately predict cancer mutations "that alter a single nucleotide in the genome" and those "that alter the order of a large stretch of the genome … such as a rearrangement, inversion, or copy-number aberration." The DREAM team is partnering with Google and Annai systems to provide the data for this particular challenge.
For the RA challenge, participants will use whole genome genotype and clinical data from 2,000 individuals with the disease who have been treated with anti-TNF therapy, to build models that predict "treatment response as measured by the change in disease activity score-28 in response to anti-TNF therapy," and to predict "poor responders" based on specific criteria.
Registration for DREAM 8.5 is now open. This round of challenges will wrap up early next summer.