The company has been using Nvidia Parabricks, a GPU-accelerated software suite, to drastically speed up genomic data analysis for DNA sequencing.
Droplet said its method can detect residual disease in 24 hours by analyzing lymphatic fluid collected post surgery, compared to the four to six weeks it typically takes for tumor remnants to appear in blood-based tests.
“By leveraging NVIDIA Parabricks’ acceleration, we’ve been able to compress some of our most computationally intensive steps from more than a day down to just a few hours,” said Droplet‘s chief scientific officer, Wendy Winckler.
The company said it also realized operational benefits despite higher hourly costs for GPU compute, adding, “The dramatically reduced runtime results in a lower overall cost per sample.”
Faster turnaround allows patients to get the results while still in the hospital, while avoiding extra visits or long waits for traditional blood tests, it added.
“We are using Parabricks to speed up our genomic analysis and shorten turnaround time from 10 days to less than five days,” said Zhuosheng Gu, senior director of informatics, R&D at Droplet Biosciences. The diagnostic startup’s first clinical test is for HPV-negative head and neck cancer, validated under the Clinical Laboratory Improvement Amendments.
Droplet is a member of NVIDIA Inception, an AI startup accelerator program, and an NVIDIA AI Enterprise customer.