By Rob Mitchum // December 12, 2014
The theme for the 2014-15 Computation Institute Inside the Discovery Cloud Speaker Series is “Catalyzing Collaboration,” placing the spotlight on the unique, multidisciplinary research partnerships enabled by the CI. As more and more research fields grapple with the linked challenges of big data and complex computational methods, the need for collaboration between computer scientists and domain scientists grows more urgent. The partnerships highlighted by this year’s series each demonstrate how merging expertise and tools can accelerate the pace of discovery and address some of today’s most pressing global issues in medicine, climate change, urban studies, and more.
For the first talk of the series, the focus was on the “new kid on the block” for data-intensive research: genomics. Since the sequencing of the human genome in 2003, scientists studying the genetics of humans and other organisms have faced an avalanche of new data, particularly as sequencing technology grows cheaper and more accessible. Though the early optimism about the ability to excavate medical treatments and knowledge from genomics faded in the face of unanticipated complexity, new statistical and computational methods restore the promise of genetic medicine.
In the first talk, CI Senior Fellow Nancy Cox, Professor and Section Chief of Genetic Medicine at the University of Chicago, talks about one such tool developed by her laboratory: PrediXcan. The algorithm uses a systems-based approach, including genomics, gene transcription, epigenomics, and clinical information, to look for genetic variants of interest for various diseases. Cox also talks about how geneticists increasingly use cloud computing to address challenges of data storage, management, and analysis.
Globus Genomics partners with the Cox lab and other groups to both provide those cloud resources and create new cloud-based tools for genetic analysis. By combining the data transfer capabilities of Globus with the elastic computation of Amazon Web Services and the analysis platform Galaxy, Globus Genomics makes it much easier for researchers to extract knowledge from their sequence data. In the second talk, Director of User Services Paul Davé provides an introduction to the service, and Biomedical Informatics Consultant Alex Rodriguez presents a new consensus variant-calling algorithm that the Globus Genomics team built in partnership with the Cox lab (previously covered here).
The next installment of the Inside the Discovery Cloud series, on December 17, will focus on parallel computing and molecular engineering, with speakers Juan de Pablo of the Institute for Molecular Engineering and Michael Wilde of Swift.