Wellcome Sanger Institute Advances Cancer Research with NVIDIA Accelerated Computing
Lawrence Jengar Sep 24, 2024 10:41
The Wellcome Sanger Institute leverages NVIDIA's accelerated computing to enhance cancer genome analysis, reducing costs and energy consumption significantly.
The Wellcome Sanger Institute, a prominent contributor to the Human Genome Project, is utilizing NVIDIA's accelerated computing technology to enhance its cancer research efforts, according to an announcement from NVIDIA Blog.
Accelerating Genomic Analysis
Based in the UK, the institute operates one of the world’s largest sequencing facilities, having read over 48 petabases of DNA and RNA sequences. This extensive data collection aids in understanding health and disease mechanisms. Specifically, the institute's Cancer, Ageing, and Somatic Mutation (CASM) program sequences and analyzes tens of thousands of cancer genomes annually to investigate mutational processes and treatment efficacy.
To manage these large-scale initiatives, the Wellcome Sanger Institute is exploring the use of NVIDIA DGX systems and NVIDIA Parabricks, a genomics analysis software suite that leverages accelerated computing to process data efficiently. Jingwei Wang, principal software developer for CASM, highlighted that NVIDIA's solutions would save considerable time, cost, and energy, enabling the institute to handle hundreds of thousands of somatic samples more effectively.
Reducing Runtime and Energy Consumption
The institute has developed high-throughput models for genome-wide functional screens and drug testing. By integrating NVIDIA accelerated computing and software, it has significantly reduced the runtime and energy consumption per genome. Specifically, employing the Burrows-Wheeler Aligner (BWA) with Parabricks on NVIDIA GPUs has resulted in a 1.6x reduction in runtime, a 24x reduction in costs, and a 42x reduction in energy consumption compared to using 128 dual-socket CPU servers.
Annually, the institute consumes about 125 million CPU hours for sequencing 10,000 genomes. By switching to BWA with Parabricks on GPUs, the institute could save approximately $1 million and 1,000 megawatt-hours of energy each year, equivalent to the energy required to power an average American home for a century.
Collaborating With Industry Leaders
The Sanger Institute's NVIDIA-accelerated sequencing lab functions as an AI factory, where data is transformed into actionable intelligence. This AI factory model leverages full-stack accelerated computing platforms to tackle computationally intensive tasks.
The institute collaborates with industry leaders like Schneider Electric to optimize data center designs for energy efficiency and high performance. Schneider Electric, known for its expertise in energy management and automation, assists in reducing data center downtime and equipping the DNA sequencing lab with essential technologies such as uninterruptible power supplies and advanced cooling systems.
During the NVIDIA GTC conference, Schneider Electric announced the release of AI data center reference designs tailored for NVIDIA accelerated computing clusters. These designs aim to enhance data processing, engineering simulation, electronic design automation, computer-aided drug design, and generative AI, focusing on high-power distribution and liquid-cooling systems.
This week, representatives from the Wellcome Sanger Institute, Schneider Electric, and NVIDIA will discuss their collaborative efforts at an NYC Climate Week panel hosted by The Economist.
For more information, visit the NVIDIA Blog.
Image source: Shutterstock