The "EasyGenomics" service offered by BGI — China's huge genomics lab in Shenzhen — uses graphics processing units (GPUs) made by U.S. tech company NVIDIA to make analysis of DNA sequencing far faster than ever. That GPU acceleration, combined with traditional computing power, reduces genomic analysis times from five days to just five hours.
"By enabling larger numbers of researchers to accelerate DNA sequencing data more easily and affordably, we hope to help facilitate the use of genomics for clinical diagnostics as a practical component of health care, as well as for complex disease research," said BGI vice President Lin Fang.
Researchers can sign up now for a limited free trial of the paid service, and BGI envisions making EasyGenomics available to everyone from biologists to physicians. The institute already does genomic sequencing for everything from the giant panda to the E. coli bacterium — all part of an effort to make better medicines, improve health care and even create genetically enhanced foods.
Fast, cheap analysis of DNA sequencing is part of the dream for personalized medicine: Saving an ill patient's life by understanding the particular genetics of that patient's disease. Such an online service could allow a physician sitting in his or her office anywhere to send a patient's genomic data and get an analysis back within hours.
The costs of genome sequencing have steadily fallen in recent years, but a lot of computing power is still needed to analyze all those DNA sequences. BGI's computing solution comes from NVIDIA's GPUs, which focus upon running many tasks efficiently — technology already used in three of the world's five most powerful supercomputers.
BGI's EasyGenomics will "get us that much closer to effective and affordable individualized treatments," said Sumit Gupta, a senior director at NVIDIA.
Online services for analyzing DNA had begun appearing even before the BGI announcement. Amazon Web Services offers supercomputing resources for researchers who want to run calculations on the world's largest database of human genetics — U.S. National Institutes of Health data from the 1,000 Genomes Project.