Element Biosciences’ AVITI platform sequences DNA using a fundamentally different detection strategy than the sequencing-by-synthesis chemistry that has dominated the field for over a decade. Instead of incorporating a labeled nucleotide into a growing DNA strand and reading the label, the AVITI system separates the act of extending the strand from the act of identifying each base, using multivalent binding molecules called avidites to achieve high accuracy with dramatically less reagent. The result is a benchtop sequencer that, across several independent comparisons, matches or closely rivals much larger instruments on data quality while operating at a different cost structure.
How Avidity Sequencing Works
The chemistry at the heart of the AVITI is called sequencing by avidity, or SBAvidity. In conventional sequencing by synthesis, a single fluorescently labeled nucleotide binds to the DNA template, gets incorporated by a polymerase, and emits a signal that a camera reads. The nucleotide does double duty: it is both the signal molecule and the building block. That coupling means the chemistry has to balance incorporation efficiency against signal brightness, and it requires relatively high concentrations of expensive labeled nucleotides floating in solution.
The AVITI decouples those two jobs. DNA extension happens first: an unlabeled nucleotide is incorporated into each growing strand by a polymerase, advancing the sequence by one base. Then, in a separate step, a probe called an avidite is washed over the flow cell to identify which base was just added. An avidite is a dye-labeled polymer with multiple identical nucleotides attached. Because the DNA on the AVITI flow cell is arranged in structures called polonies, which contain many concatemer copies of the same fragment, one avidite can grab onto several complementary sites at once. That multivalent grip is the “avidity” in the name, and it produces an extremely stable complex while the camera images the flow cell.1Nature Biotechnology. Sequencing by avidity enables high accuracy with low reagent consumption
After imaging, the avidite is washed away under specific conditions, leaving no chemical modifications on the synthesized strand. This is a meaningful difference from older reversible-terminator approaches, where each cycle leaves a small scar from cleaving the terminator group. Because the avidite never gets incorporated, the growing strand stays chemically clean cycle after cycle.
Why Separating Extension from Identification Matters
Splitting the sequencing process into two independently optimizable halves has a few practical consequences. First, the polymerase can be tuned purely for accurate, efficient incorporation without worrying about how a bulky fluorescent tag might slow it down or introduce errors. Second, the identification step can be optimized for signal strength and specificity on its own terms. Multiple fluorophores on a single avidite polymer boost the signal-to-background ratio, making it easier to distinguish genuine signal from noise during imaging.2PubMed Central. Utility analyses of AVITI sequencing chemistry
The multivalent binding also slashes reagent consumption. Because each avidite molecule can engage multiple complementary sites simultaneously, the system needs roughly 100-fold lower concentrations of reporting nucleotides compared to single-nucleotide binding approaches.1Nature Biotechnology. Sequencing by avidity enables high accuracy with low reagent consumption In practical terms, that translates to cheaper consumable kits and less chemical waste per run. For labs sequencing at moderate to high volumes, consumable cost is often the single largest line item in a sequencing budget, so even modest per-run savings compound quickly.
The tight binding between avidite and template also means dissociation during imaging is negligible, so the signal does not decay meaningfully while the camera is recording. That stability contributes to cleaner raw data with lower background, which feeds directly into higher base-call quality scores.
How AVITI Data Quality Compares to Illumina Platforms
Because Illumina instruments have been the default for short-read sequencing in most labs, the natural question is how AVITI data stacks up. Several independent groups have now run matched libraries on both platforms and published side-by-side results.
In one comparative study, matched DNA and RNA libraries were sequenced on the AVITI and Illumina’s NextSeq 550. For PCR-free DNA libraries, the AVITI produced roughly 90% lower experimentally determined error rates. For short-read RNA quantification, the error rate was about a third lower on the AVITI.3PubMed Central. Utility Analyses of AVITI Sequencing Chemistry Those are large margins, though it is worth noting the comparison was against the NextSeq 550, a workhorse instrument but not Illumina’s newest or highest-end platform. When the comparison shifts to more recent Illumina hardware, the gap narrows.
A whole-genome sequencing study comparing the AVITI directly to Illumina’s NovaSeq X Plus found that variant-calling performance was highly comparable between the two platforms. The AVITI showed slightly higher accuracy for insertions and deletions at low sequencing depth, but at standard coverage the instruments were essentially neck and neck.4NAR Genomics and Bioinformatics. Whole-genome sequencing with AVITI and NovaSeq X Plus reveals comparable performance with contextual biases That study also noted that each platform carried its own context-dependent biases, meaning certain sequence motifs were slightly harder for one chemistry than the other. Neither platform was universally better across every genomic context.
A broader benchmarking effort that tested nine short-read chemistries across seven different sequencers placed the AVITI’s UltraQ chemistry among the platforms with the lowest substitution error rates overall.5bioRxiv. Short-Read Sequencing Benchmarking with Donor-Specific Assemblies Substitution errors are the most common type of mistake in short-read sequencing, so performing well on that metric matters for most downstream applications. The same benchmarking found that different platforms had different weaknesses, with some chemistries more prone to insertion and deletion artifacts in specific sequence contexts, so the best choice of sequencer can depend on what kind of errors are most consequential for a given experiment.
Clinical and Diagnostic Applications
For clinical laboratories, a sequencer needs to reliably detect medically meaningful variants, not just produce low aggregate error rates. A study focused on hematological malignancies ran the same tumor samples through both the AVITI and Illumina’s NextSeq 550 using standard clinical assays. All 105 single-nucleotide variants and all 39 gene fusions identified on the NextSeq were also detected by the AVITI, and the correlation in variant allele frequencies between the two platforms was extremely high.6The Journal of Applied Laboratory Medicine. Comparison of Sequencing-by-Synthesis and Avidity Base Chemistry Next-Generation Sequencing Platforms in Identifying Somatic Variants of Hematological Malignancies That degree of concordance is reassuring for labs considering the AVITI for cancer panel testing, because it suggests existing validated assays can migrate to the new platform without sacrificing sensitivity.
Clinical sequencing is a context where the AVITI’s lower reagent costs could be especially impactful. Many hospital genomics labs run moderate volumes of targeted panels rather than high-throughput whole genomes. For that workload, a benchtop instrument with affordable consumables can be more practical than a high-capacity production sequencer that needs large batches to be cost-efficient. The AVITI’s flow cell design, which supports flexible run configurations, fits that use case reasonably well.
Genome-Wide Genotyping and Agricultural Genomics
Genotyping-by-sequencing, where you sequence a reduced representation of many genomes rather than deeply sequencing one, is heavily used in agriculture, conservation biology, and population genetics. A study that sequenced and genotyped 40 cannabis samples on both the AVITI and Illumina NovaSeq found strong overlap: after filtering, about four-fifths of variants were shared between the two data sets, and where both platforms called the same variant, genotype calls agreed nearly 99% of the time.7Genome. AVITI as an alternative to Illumina for low-cost genome-wide genotyping
The roughly 19% of variants that were unique to one platform or the other largely reflected coverage differences and filtering choices rather than fundamental chemistry failures. For breeding programs and genetic mapping studies, where relative genotype accuracy matters more than capturing every last rare variant, that level of concordance is good enough to treat the AVITI as a viable alternative. The appeal is straightforward: if a plant-breeding lab can sequence the same number of samples for less money per sample, it can either save budget or increase sample sizes, and larger sample sizes translate directly into more statistical power for detecting quantitative trait loci.
Microbiome and Metagenomics
Microbiome studies have traditionally relied on short amplicon reads targeting one or a few variable regions of the bacterial 16S ribosomal RNA gene. That approach works for broad community profiling but often cannot resolve closely related species, let alone strains. A study of the soybean root microbiome used synthetic long reads on the AVITI to reconstruct full-length 16S sequences spanning all nine hypervariable regions, as well as the full eukaryotic rRNA operon spanning the 18S, ITS1, and ITS2 regions.
The results were striking. Every full-length prokaryotic amplicon sequence variant was resolved to the species level, and nearly a quarter reached strain-level identification. In the eukaryotic fraction, which is often neglected in microbiome work because short reads cannot distinguish many protists and fungi, the full-length approach identified organisms across five kingdoms spanning 19 genera of protists in addition to fungi.8PubMed Central. Fine-scale characterization of the soybean rhizosphere microbiome via synthetic long reads and avidity sequencing Thirteen species of the nitrogen-fixing genus Bradyrhizobium were distinguished, with strain-level calls for the two species most commonly associated with soybean nodulation. That resolution is far beyond what standard short-read 16S surveys achieve.
This kind of fine-grained microbial profiling matters in agriculture, where knowing which specific strains colonize a plant’s root zone can inform decisions about soil management and inoculant selection. It also matters in clinical metagenomics, where distinguishing a pathogenic strain from a closely related commensal can change treatment decisions. The ability to get strain-level data from a benchtop sequencer, rather than requiring either expensive long-read instruments or complex computational assembly pipelines, lowers the barrier for labs that want higher taxonomic resolution without overhauling their infrastructure.
Synthetic Long Reads on a Short-Read Platform
The soybean microbiome study hints at a broader point worth unpacking. The AVITI is a short-read instrument, typically generating reads of a few hundred bases. But through library preparation methods that tag and link fragments from the same original long molecule, it can reconstruct sequences thousands of bases long after the fact. These are called synthetic long reads, and they bridge the gap between the low cost per base of short-read sequencing and the structural information of native long-read platforms.
In the matched-platform comparison that tested synthetic long-read RNA sequencing, both the AVITI and Illumina’s NovaSeq 6000 performed comparably in quantifying genes and isoforms, with the AVITI showing a marginally lower error rate for the long-read data and fewer chemistry-specific errors.3PubMed Central. Utility Analyses of AVITI Sequencing Chemistry For labs interested in transcript isoform detection or structural variant calling without purchasing a dedicated long-read sequencer, synthetic long reads on the AVITI represent a pragmatic middle path.
Functional Genomics and Pooled Screens
One of the more creative applications emerging for the AVITI sits at the intersection of imaging and sequencing. A recently described platform directly sequences single guide RNAs and endogenous 3′ untranslated regions in fixed cells while simultaneously measuring protein abundance and cellular morphology. Researchers used this approach to perform an optical pooled screen of CRISPR-perturbed lung cancer cells, linking genetic perturbations to phenotypic outcomes within a single experiment.9bioRxiv. Direct In-Sample Sequencing of the 3′ Transcriptome Expands the Capabilities of Optical Pooled Screens
Optical pooled screens are a growing area in functional genomics. The traditional approach is to separate the sequencing step (which tells you what perturbation each cell received) from the phenotyping step (which tells you what happened to the cell), then computationally match the two. Performing both in the same sample reduces batch effects and information loss. The AVITI’s chemistry, with its low background signal and stable probe binding, lends itself to the kind of in situ readouts these experiments demand.
Platform-Specific Biases and Practical Trade-offs
No sequencing chemistry is uniformly accurate across all DNA sequences, and the AVITI is no exception. The whole-genome comparison against the NovaSeq X Plus found that each platform had its own contextual biases: certain GC-rich or repetitive motifs were harder for one chemistry than the other.10PubMed Central. Whole-genome sequencing with AVITI and NovaSeq X Plus reveals comparable performance with contextual biases The nine-platform benchmarking study reinforced this point, showing that different chemistries had different error profiles even when their aggregate error rates were similar.5bioRxiv. Short-Read Sequencing Benchmarking with Donor-Specific Assemblies
What this means in practice is that switching platforms is not as simple as swapping one sequencer for another and expecting identical results at every locus. Bioinformatics pipelines tuned to Illumina error profiles may need recalibration. Variant callers optimized for one error distribution may lose sensitivity or specificity when encountering a different one. Labs migrating validated assays should run concordance studies with their own panels, as the hematological malignancy study did, before relying on new data for clinical decisions.
Throughput is another practical consideration. The AVITI is a benchtop instrument. It is not designed to compete with the NovaSeq X Plus on raw output per run. For large-scale population sequencing projects generating tens of thousands of whole genomes, high-throughput production instruments still make more sense. The AVITI’s sweet spot is labs that need high accuracy at moderate scale: clinical panels, targeted assays, amplicon sequencing, smaller whole-genome projects, and applications where consumable cost per sample matters more than maximum daily throughput.
Where the Competitive Landscape Stands
The short-read sequencing market has historically been dominated by a single vendor, which made it difficult for labs to negotiate pricing or diversify their instrument risk. The AVITI, along with platforms from other newer entrants, has changed that dynamic. Having a second viable short-read option gives core facilities and clinical labs leverage they did not have five years ago, and it creates competitive pressure that benefits buyers regardless of which platform they ultimately choose.
Element Biosciences has also continued developing the platform. The UltraQ chemistry mentioned in the nine-platform benchmarking represents an update to the original AVITI chemistry, pushing substitution error rates even lower. For a platform that launched relatively recently, the pace of chemistry iteration matters because it signals that the accuracy ceiling has not been reached and that the underlying approach has room to improve further. Whether the AVITI eventually matches, exceeds, or simply stays competitive with future Illumina chemistries is an open question, but the early data suggests it has earned a place in the conversation rather than being a curiosity that labs can safely ignore.