By EVOBYTE Your partner in bioinformatics
Introduction
Whole‑genome sequencing used to take days. In 2025, a Broad Clinical Labs team, working with Roche and Boston Children’s Hospital, pushed that down to under four hours, earning a Guinness World Records title for the fastest DNA sequencing technique. The chemistry behind the sprint was Sequencing by Expansion, or SBX.
What is Sequencing by Expansion (SBX)?
SBX is Roche’s new single‑molecule method that first “rewrites” each DNA template into a longer surrogate called an Xpandomer. The surrogate embeds the original base sequence using engineered building blocks (X‑NTPs) and then threads through a proprietary nanopore sitting on a high‑throughput CMOS sensor. Because the Xpandomer is roughly 50× longer and carries high signal‑to‑noise reporters, the electrical readout becomes cleaner and faster than reading native DNA directly. In early evaluations, Roche has reported flexible read lengths from about 50 to over 1,000 bases, whole‑genome F1 accuracy above 99.8% for SNVs on HG001, and sample‑to‑VCF workflows in hours rather than days.
How SBX differs from Illumina SBS and Oxford Nanopore
Illumina’s sequencing by synthesis (SBS) images fluorescent, reversible terminators as DNA is copied base‑by‑base on dense flow cells. It delivers very high accuracy with short reads and enormous throughput, and remains the most widely deployed NGS platform. SBX, by contrast, is single‑molecule and electrical: it measures current changes as the expanded surrogate passes a nanopore on a CMOS chip. The intent is to pair nanopore‑style real‑time sensing with short‑to‑mid read lengths and elevated accuracy via the Xpandomer intermediates.
Oxford Nanopore also senses ionic current through pores, but it reads native DNA or RNA directly, enabling ultra‑long reads and direct methylation/RNA detection. SBX uses nanopore detection too, yet it sequences the engineered Xpandomer rather than the original strand, trading native‑molecule access for speed, signal quality, and tunable read lengths.
What the output data looks like
Today’s SBX runs produce familiar bioinformatics artifacts. Basecalls arrive as FASTQ with per‑base quality scores; secondary analysis yields BAM/CRAM alignments and VCFs. In demonstrations, teams have shown rapid end‑to‑end WGS—alignment and variant calling included—within a single work shift. Practically speaking, if you receive FASTQs from an SBX run, your standard short‑read pipeline will feel natural:
minimap2 -ax sr ref.fa reads.fq.gz | samtools sort -o aln.bam
bcftools mpileup -Ou -f ref.fa aln.bam | bcftools call -mv -Oz -o calls.vcf.gz
Roche also highlights “duplex” modes that link complementary strands for higher consensus accuracy, plus workflows exploring methylation mapping and RNA applications—signals that the data types will broaden beyond vanilla DNA variant calling.
Market snapshot in 2025
As of December 1, 2025, SBX remains in development and is not yet commercially available; usage is concentrated in early‑access collaborations at institutions such as Broad Clinical Labs and the Wellcome Sanger Institute. Roche has previewed its AXELIOS system around SBX, with industry reporting pointing to a 2026 commercial debut. Meanwhile, Illumina’s SBS platforms continue to dominate installed base and data output, and Oxford Nanopore maintains leadership in native long‑read sequencing and field‑portable devices. Expect SBX adoption to start in rapid WGS and hybrid short/long‑read workflows once systems and kits become broadly available.
Summary / Takeaways
Sequencing by Expansion turns DNA into an easier‑to‑read surrogate before nanopore detection, aiming to blend real‑time speed with strong accuracy and flexible read lengths. If you know Illumina pipelines, SBX data will slot into familiar FASTQ‑to‑VCF workflows, yet it may open doors to faster triage, richer isoform analysis, and streamlined methylation assays. The market is watching closely; 2026 should reveal how SBX performs outside pilot sites and how it reshapes price, speed, and read‑length trade‑offs across NGS.