Inova Powers Genomics with SGI

Print Friendly, PDF & Email

inovaToday SGI announced that Inova Translational Medicine Institute (ITMI) has selected the company’s supercomputing solutions to enable researchers to obtain insights from its premier genomic databases. With the ability to diagnose patients with more accuracy and speed, ITMI will enable a higher level of treatment and care for the diverse population it serves.

Deploying the SGI UV 2000 system into our infrastructure has been invaluable to our research initiatives, which have greatly improved since installing the system,” said Dr. Joe Vockley, chief operating officer and chief scientific officer at ITMI. “Additionally, the ease and level of support around the system installation got us up and running quickly which greatly helped us adapt to a new high-performance computing environment.”

ITMI’s genomic database is one of the largest in the world. As part of the Inova not-for-profit healthcare system based in Fairfax, VA, ITMI works to transform healthcare from reactive to predictive medicine by analyzing and modeling massive amounts of data from its genome database, which is one of the largest in the world and the only that is connected directly to a community health system. Researchers at ITMI seek to understand the causes of diseases and rare disorders, how they develop and what the best treatments are to improve patient care.

SGI_logo_platinum_lgIn order to improve the practice of personalized and precision medicine by leveraging the genome sequences of healthy and sick individuals, ITMI requires significant compute power. To handle the workload, ITMI selected the SGI UV 2000 advanced symmetric multiprocessing (SMP) system using Intel Xeon E5-4650L for the 512 cores with 32 blades and four individual rack units. The system also includes the SGI InfiniteStorage 5600, with 50 gigabytes (GB) per second of input/output (I/O), providing one petabyte of storage capacity in a single file system, making this system SGI’s most I/O intensive UV machine. The system provides best-in-class memory-to-processor ratio, at 32 GB per real core or 16 GB per hyper threaded core, automating the processing of frequently read data and leaving large amounts of memory available to speed reading files, increasing the efficiency of data processing and analysis.

ITMI has conducted three core studies for which it employs the compute power of the SGI UV 2000, including:

  • Pre-term birth study of 1,000 families to find biological and clinical features to help predict and prevent pre-term births.
  • Longitudinal birth study, with the goal of enrolling between 5,000 to 10,000 expecting families (~25 thousand genomes) and tracking the infant over the course of 18 years to find correlations between disease, genomics, patient demographics and environmental factors.
  • Translational study that focuses on babies in the neonatal intensive-care unit (NICU) who are sick and not yet diagnosed. Using their genomic database, ITMI doctors use whole genome sequences for those infants and family members, generating successful diagnosis of sick infants greater than 60 percent of the time.

In addition to delivering high performance for compute-intensive workloads, ITMI’s UV 2000 system also simplifies management by enabling researchers to develop and use their own code, rather than having to adapt to a pre-designated one. This lowers the Institute’s administrative IT burden while simultaneously increasing research workflow capabilities, enabling ITMI to devote greater resources to the challenging chronic disease landscape of the diverse population of families they serve.

SGI is pleased to partner with ITMI once again to provide the SGI UV 2000 system for its critically important genomic research and modeling,” said Jorge Titinger, president and CEO, SGI. “By running these compute intensive workloads with greater accuracy and speed, ITMI can gain new insights that dramatically improve patient care.”

The SGI UV 2000 system at ITMI is currently in use to support the Institute’s research and promote patient care.

Sign up for our insideHPC Newsletter.