Transforming Cancer Research with a New Network Model
Curing cancer took a leap forward in 2003, when the Human Genome Project completed its goal of identifying and mapping the human genome. Since then, the genomic knowledge base has expanded exponentially, creating data files that are hundreds of gigabytes of data in size. To make this data more widely available, the Ontario Institute for Cancer Research (OICR), a translational cancer research institute based in Ontario, Canada, launched an academic research cloud known as the Cancer Genome Collaboratory. The Collaboratory houses the data of the International Cancer Genome Consortium—a global collaboration involving more than 70 projects and 40 countries—to sequence the genomes of 25,000 tumors and their matched normal tissues across 50 major cancer types. Collaboratory users have fast, easy access to this unique data set.
Before the cloud, scientists conducted research on High-Performance Computing (HPC) clusters. These powerful resources shared storage across large numbers of servers, making collaboration possible—but not easy. Scientists had to send and retrieve gigantic files. File transfers could take weeks, which was costly and slowed research progress, and scientists were often limited in their technology choices for developing their analysis algorithms.
Today’s genome sequencing computers generate so much data that storage requirements and costs rise much faster than organizations can afford to support. It’s not financially feasible to have identical data sets stored in multiple places. More research environments are moving to cloud computing, which allows users to create the precise environments they need for their experiments and to use the latest technologies.