This presentation provides an update on the Stratos Genomics Sequencing by Expansion (SPX) technology, combining its novel chemistry with a Genia high-throughput nanopore sensor array. The core innovation involves sequencing "expandimers," modified DNA surrogates, to enhance signal-to-noise ratios. The talk highlights significant improvements in data quality and speed, featuring updated F1 scores for both Simplex and Duplex sequencing approaches, which have achieved Q40 average base quality and sub-5-hour turnaround from sample to VCF. New data is presented for challenging applications, including FFPE (Formalin-Fixed Paraffin-Embedded) tissue samples and Minimal Residual Disease (MRD) detection, demonstrating superior accuracy and coverage uniformity compared to existing technologies. The platform's flexibility for varying read lengths and throughput is emphasized, with initial data for longer-read Simplex applications showcasing its potential for genomic phasing and variable number tandem repeat (VNTR) analysis.
Exploring Stratos Genomics Sequencing by Expansion (SPX): Updates and New Applications
0:02 Introduction & Background: The presentation builds on previous webinars, a seminal preprint covering chemistry up to 2020, and the AGBT technology launch. It aims to quickly review the technology overview to focus on new data.
1:27 Foundational Principles: The SPX approach is a single-molecule sequencing technology designed for flexibility (short to long runs, varying throughput) and high performance (accuracy, throughput, cost efficiency).
2:00 Performance Metrics: The technology boasts 500 million bases per second throughput.
2:19 Simplex vs. Duplex Reads: Simplex reads offer longer reads and higher throughput at Q20 accuracy, while Duplex reads provide higher accuracy.
2:30 Rapid Sequencing (SPX Fast): The SPX fast approach achieves sample-to-VCF in less than 5 hours, an improvement from 6 hours 25 minutes to 4 hours 23 minutes (excluding initial DNA prep, which adds ~20 minutes for a total of ~4 hours 43 minutes from blood sample).
3:19 Technology Integration: SPX combines Stratos Genomics' Sequencing by Expansion chemistry with Genia's high-throughput sensor array. Synthesis occurs on a separate instrument, followed by sequencing, offering workflow flexibility.
4:00 Expandimer Chemistry: The core innovation involves "expandimers," synthetic DNA surrogates where each nucleotide is "expanded" with a reporter code, translocation control element, and enhancers to improve signal-to-noise in nanopore measurements.
6:09 Expandimer Synthesis: Expandimers are synthesized on a chip through template-dependent copying of DNA, followed by two orthogonal cleavage steps to release a clean molecule for nanopore analysis.
6:55 Nanopore Measurement & Translocation Control: A key feature is precise translocation control within the nanopore, allowing for statistical sampling at a base voltage and rapid, microsecond-level advancement to the next base, ensuring deterministic throughput modulation and high measurement quality.
8:41 High-Throughput Array: The technology has transitioned from single-pore systems to an 8-million pore array, combining microwells, electrodes, and detection circuits for massive throughput.
9:10 Sensor Module & Reusability: The sequencing instrument uses a sensor module with the 8-million pore chip, enabling fluidic control and a cyclical workflow where the system can be cleaned and reused, which is critical for cost reduction.
10:00 Pore Activity & Durability: The system typically starts with 7+ million sequence-quality pores, maintaining around 5 million functional pores over a 4-hour run.
10:36 Signal Resolution: High signal resolution is maintained on the 8-million array, allowing clear distinction of individual expandimer molecules and their base calls.
11:09 Flexible Multiplexing: The system supports multiplexing (e.g., 1 or 4 expandimer pools) and flexible sequencing times (4 minutes to 4 hours).
11:39 Duplex Sequencing (SPXD): Utilizes a Y-adapter and hairpin library structure for intramolecular consensus, yielding very high accuracy.
12:00 Duplex Applications: Applied to MRD (Minimal Residual Disease), rapid Whole Genome Sequencing (WGS), and FFPE (Formalin-Fixed Paraffin-Embedded) samples.
12:21 SPXD Workflow: Involves library fragmentation, ordered ligation with Y-adapter/hairpin, linear amplification (unique for accuracy), and expandimer synthesis.
13:29 Accuracy Update (Genome-in-a-Bottle): For 1-hour runs, achieving 5 billion duplex reads with Q40 average duplex base quality. F1 scores for SNVs and indels have improved, with 99.9% coverage at >10x. Good performance up to 17bp homopolymer regions.
16:47 SPX Fast Workflow: An amplification-free workflow that uses more sample (e.g., 2 micrograms) to achieve faster results and even better homopolymer accuracy (up to 21bp).
20:19 Concurrent Analysis: A significant speed improvement comes from concurrent demultiplexing, intramolecular consensus mapping, and alignment during sequencing, eliminating hours of analysis time through accelerated algorithms (e.g., with Nvidia's Parabricks libraries).
21:26 FFPE Data (New): First-ever FFPE data using SPXD workflow with DNA repair steps.
22:20 FFPE Comparison with Illumina: SPX showed lower SNV/indel error rates, higher long homopolymer accuracy, and better coverage uniformity in high GC regions compared to Illumina sequencing, with ~10% better recall for SNVs/indels in non-difficult regions. Good concordance for copy number variants.
26:08 MRD Data (New): Achieved Q50 accuracy by combining consensus reads, super consensus collapsing, and filtering.
27:21 MRD Results: Successfully detected positive MRD in all 15 cancer samples tested (4 nanograms DNA each), with an estimated tumor fraction of 10^-6.
SBXS (<500bp): Higher throughput for applications like 10x Flex (14 billion cell ranger valid reads in 1 hour).
SPXSL (>500bp): Focuses on longer reads for RNA isoform analysis, genomic DNA phasing, and Variable Number Tandem Repeats (VNTRs).
30:00 SPXSL for Phasing/VNTRs: With a billion reads at ~1000 base length in one hour, SPXSL enables detection of VNTRs and phasing of alleles that shorter-read technologies (like Illumina or SPXD) cannot. Shows dramatic improvement in F1 scores for longer reads.
32:39 Pathogenic Variant Detection: SPXSL correctly detected a pathogenic compound heterozygous variant associated with hyperphenylalaninemia, which was missed by Illumina and SPXD.