Ultima Genomics, Inc. unveiled a broad collaboration with NVIDIA that leverages AI-accelerated analysis across multiple stages of the DNA sequencing and analysis process. The collaboration includes the use of NVIDIA GPUs for deep learning in photometry and base calling and will soon include acceleration of variant calling.
Over the last five years, Ultima Genomics has developed a fundamentally new sequencing architecture designed to scale beyond conventional approaches, including completely different flow cell engineering, sequencing chemistry, and machine learning. Ultima is currently in an early access program for the UG 100
Genomics Meets AI
The Ultima instrument utilizes NVIDIA A30 GPUs with the NVIDIA Clara AI computing platform to offer highly accurate and accelerated primary and secondary analysis for customers who want to run genomic analyses at scale. Ultima has separately announced partnerships to develop Ultima-specific germline variant calling pipelines for Google’s DeepVariant. These custom models can be run within NVIDIA Clara Parabricks, which accelerates these and other applications for secondary analysis.
“Ultima’s mission is to continuously drive the scale of genomic information,” said Gilad Almogy, CEO of Ultima Genomics. “Our unique architecture scales by combining multiple technologies, so we are excited to work with NVIDIA, leveraging their platforms for GPUs and AI acceleration, to power high-quality sequencing and accelerate the computing to lower total cost of ownership for customers.”
“NVIDIA solutions are designed to accelerate high-volume data processing, and our collaboration with Ultima demonstrates this at multiple levels,” said George Vacek, Global Director of Genomics Alliances at NVIDIA. “NVIDIA GPUs provide a high degree of onboard computational power within the UG 100, and data analysis can be accelerated downstream via Clara Parabricks, enabling large-scale data initiatives via analysis that is both higher throughput and lower cost.”
NVIDIA Clara Parabricks computational pipelines employ NVIDIA’s CUDA, HPC, AI and data analytics stacks to deliver GPU-accelerated libraries, pipelines, and reference application workflows for primary, secondary, and tertiary analysis. Off-the-shelf solutions are coupled with a toolkit to support new application development to address the rapidly evolving needs of high-throughput genomic labs.
