Revio from Pacific Biosciences: Latest Sequencing Technologies—Part 3

Pacific Biosciences (PacBio) revealed two new sequencers during this year’s American Society of Human Genetics meeting. The first instrument, Revio, which will be covered in this article, is a significant upgrade from their previous Sequel II device. It has lower costs, shorter run times, more SMRT Cells and zero-mode waveguides (ZMW), and significantly higher output compared to PacBio’s other sequencers.

Image of Revio (courtesy of PacBio)

Read type and length
As a part of PacBio’s long-read sequencing systems, Revio was designed to perform their signature High Fidelity (HiFi) sequencing. The read lengths can vary from 15–18 kb with a read accuracy of 99.95% (Q33).

Flow cell/SMRT Cell options
Revio supports 4 high-density SMRT Cells that can be run in parallel or independently. Each SMRT Cell contains 25 million ZMWs and the run times have been reduced to 24 hours. This is a significant increase from the Sequel IIe, which ran a single SMRT Cell that only supported 8 million ZMWs for a 30-hour sequencing run. The upgrades in SMRT Cells, ZMWs, and run times account for a 15x increase in data generation.

Data output
The data output of Revio is up to 90 Gb per SMRT Cell or a total of 360 Gb of HiFi reads per day. Annual outputs are estimated for 1,300 Revio SMRT Cells per year: the equivalent of 1,300 human genomes at 30X coverage. HiFi reads are outputted in BAM format, and the output files are greatly reduced in size (by 50% per base) despite the increase in data generation.

Analysis options
Revio comes equipped with NVIDIA GPUs for a >20x higher computing power than their previous instruments. Demultiplexing is completed on the sequencer as well as direct methylation calling. In addition, Google teamed up with PacBio to produce Google Health DeepConsensus, which has been included in the instrument. Google Health DeepConsensus uses deep-learning technology to correct for errors during sequencing and leads to an increased data yield and base calling accuracy¹.

Common applications
• Variant calling (SNVs, indels, SVs)
• Whole genome sequencing
• Single-cell transcriptomics
• Large gene panels
• 5mC at CpG sites

Release date
PacBio is already taking orders for Revio and shipments are expected to begin in the first quarter of 2023.

Highlights
• Shorter run times with significantly increased (15x) data output and an increased HiFi read throughput
• Easier setup and flexibility—no nitrogen is required, there are fewer consumables, and runs can be queued during sequencing.
• Long HiFi reads (>10 kb) can be used to resolve small variants, structural variants, and repeat expansions while providing uniform coverage
• An estimated $1,000 complete, phased genome
• Enhanced Google Health DeepConsensus algorithms decrease basecalling errors and increase data yields.

References
1. Baid, G., et al. (2022). DeepConsensus improves the accuracy of sequences with a gap-aware sequence transformer. Nature Biotechnology, 1-7.

To compare sequencing technologies please visit Biocompare