Researchers from Stanford University, NVIDIA, Oxford Nanopore Technologies, Google, Baylor College of Medicine and the University of California at Santa Cruz have developed a method to do DNA sequencing in 5 hours and 2 minutes – and entered the Guinness World Record for fastest DNA sequencing technique.
The research team, led by Stanford University, used AI to expedite the end-to-end process, from collecting a blood sample to sequencing the whole genome and identifying variants linked to diseases. The researchers made the diagnosis for a three-month-old infant suffering from a rare seizure-causing genetic disorder in a few hours. The traditional gene panel analysis takes as long as two weeks to return results.
By optimising the diagnosis pipeline at 7-10 hours, clinicians can quickly identify genetic clues to inform patient care plans. In this pilot project, the genomes were sequenced for 12 patients – most of them children–at Stanford Health Care and Lucile Packard Children’s Hospital Stanford.
The researchers optimised the pipeline including speeding up sample preparation and using nanopore sequencing on Oxford Nanopore’s PromethION Flow Cells to generate over 100 gigabases of data per hour. The data was then sent to NVIDIA Tensor Core GPUs in a Google Cloud computing environment for base calling. At this stage, raw signals from the device are turned into a string of A, T, G and C nucleotides, and alignment in near real-time. Since it distributed the data across cloud GPU, it instantly helped minimise latency.
The next step was to find tiny variations in the DNA sequence that can cause a genetic disorder. This stage was sped up with Clara Parabricks using a GPU-accelerated version of PEPPER-Margin-DeepVariant, a pipeline developed in a collaboration between UC Santa Cruz’s Computational Genomics Laboratory and Google. For highly accurate variant calling, DeepVariant uses convolutional neural networks. The GPU-accelerated DeepVariant Germline Pipeline software in Clara Parabricks provides results at then times the speed of native DeepVariant instances, decreasing the time to identify disease-causing variants.
The details of the ultra-rapid sequencing method is published in the New England Journal of Medicine.
