Genomics Core
Efficient and reliable analyses utilizing ai capabilities
The UK Genomics Core facility utilizes a suite of AI-enhanced software tools that increase the speed and reliability of genetics-based analyses. Specifically, clinical exome sequencing has been optimized using NVIDIA Parabricks GPU-accelerated bioinformatics software to align raw reads generated from a sequencing machine (FASTQ file) to a reference genome (BAM file) and call variants using another GPU-accelerated workflow (Google’s DeepVariant). The completely automated Parabricks pipeline effectively reduces the time it takes to find genetic variants from several hours to a few minutes. Additional AI-enhanced filtering and annotation steps are currently used in clinical workflows to provide pathologists with timely and accurate genetic data. The Genomics Core facility actively engages in research to improve disease risk prediction using AI and genomic large language models (e.g., the Nucleotide Transformer).
What is NVIDIA Parabricks?
Software Overview
NVIDIA Parabricks delivers major improvements in throughput time for common analytical tasks in genomics, including germline and somatic analysis. The core of the Parabricks software is its tight integration with the GPU, which takes raw data and transforms it according to the users requirements.
Accelerated Genomics Analysis
NVIDIA Parabricks works with open-source tools to provide GPU-accelerated versions for secondary analyses.
Compatibility
Compatible with leading sequencing instruments, researchers, bioinformaticians, and data scientists can use Parabricks across diverse bioinformatics workflows.
Increases Efficiency
Parabricks increases speed with faster runtimes and reduces cost—while enhancing accuracy and ensuring transparency.
For more information, visit NVIDIA Parabricks.
Projects Supported by Genomics Core
