Illumina releases SpliceAI2 to help advance rare disease research

Researchers can access SpliceAI2 through Illumina's BioInsight applications, including DRAGEN Annotation and Emedgene.

Rhea-AI Summary

Illumina (ILMN) introduced SpliceAI2, a genomic AI model designed to help researchers identify disease-relevant genetic variants that affect RNA splicing. Splicing assembles RNA transcripts before protein production; changes can disrupt gene and protein function.

A newly released preprint evaluated the model against the original SpliceAI and alternatives. In matched DNA and disease data from Genomics England cohorts, SpliceAI2 identified 17% more disease-relevant splice variants than other models. Analysis of matched whole-genome and RNA sequencing data from the NIH's Genotype-Tissue Expression Portal showed 34% improved quantification of splice-site usage versus the next best model.

SpliceAI2 was trained on a dataset 100 times larger than the original model's. Illumina says its suite of SpliceAI2, PromoterAI and PrimateAI-3D enables identification of up to twice as many variants with predicted biological impact.

News Explained

Illumina says customers can access SpliceAI2 through its BioInsight applications, including DRAGEN Annotation and Emedgene, identifying a current route for researchers to use the newly introduced model.

Argus 15 min delay 4 alerts

-4.12% vs previous close $256.72 last price 1.1x rel. volume Open Argus

Details

Market Reaction – ILMN

$255.51 – $267.74 Day Range

$38.76B Market Cap

On Oct 8, the day this news came out, the latest delayed price for ILMN is 4.12% below the previous close. Our momentum scanner has recorded 4 alerts for this stock so far that day. The latest delayed price is $256.72.

Data tracked by StockTitan Argus (15 min delayed). Upgrade to Gold for real-time data.

Key Figures

Disease-relevant splice variants identified: 17% more Splice site usage quantification: 34% improvement Variants with predicted biological impact: Up to 2x as many +1 more

Disease-relevant splice variants identified

17% more

Rare disease research dataset; compared with other models

Splice site usage quantification

34% improvement

Matched whole genome and RNA sequencing data; versus the next best model

Variants with predicted biological impact

Up to 2x as many

Collectively enabled by Illumina's genomic AI models

Training dataset size

100 times larger

SpliceAI2 training dataset compared with the original SpliceAI

Key Terms

variants of uncertain significance, missense variants, cryptic splice sites, rna sequencing

4 terms

variants of uncertain significance medical

"resolving variants of uncertain significance in rare disease research"

Variants of uncertain significance are genetic changes discovered during DNA testing for which scientists do not yet know whether they cause disease, are harmless, or have no effect. For investors in biotech, diagnostics and healthcare, these ambiguous results matter because they can limit how useful a genetic test or a targeted therapy is, affect regulatory labeling and reimbursement, and add uncertainty to a company’s commercial prospects — like an unclear weather report making planning and valuation harder.

missense variants medical

"including splice, promoter, and missense variants"

A missense variant is a single-letter change in a gene’s DNA sequence that alters one codon so it specifies a different amino acid in the resulting protein. Because proteins’ shapes and activities depend on their amino acid sequence, a missense variant can change how a protein folds, binds or functions — sometimes with no effect, sometimes reducing or altering activity, and occasionally causing disease or creating a new trait. Missense variants are distinct from silent (no amino-acid change) and nonsense (creates a stop codon) mutations; whether a specific missense change matters depends on which amino acid is substituted and where it occurs in the protein.

cryptic splice sites medical

"Cryptic splice sites contribute to a significant burden of rare disease cases"

Sequences in pre-mRNA that resemble the normal (canonical) splice donor or acceptor motifs and can be used by the cellular splicing machinery to cut and join RNA when the usual splice sites are weak or disrupted. Cryptic splice sites are not the intended boundaries between exons and introns; when they are activated by genetic variants or changes in splicing regulation, they produce abnormal messenger RNA transcripts—for example causing partial exon inclusion, exon skipping, intron retention, frameshifts, or introduction of premature stop codons—which can alter the encoded protein or trigger RNA decay. The key distinction from canonical splice sites is that cryptic sites are latent, lower‑strength motifs that become functionally used only under altered sequence or regulatory conditions.

rna sequencing technical

"matched whole genome and RNA sequencing data"

RNA sequencing is a laboratory method that reads the active genetic messages inside cells, like scanning the recipe cards a cell is using at a given moment to make proteins. For investors, it matters because the results can reveal how diseases operate, identify targets for new drugs or diagnostics, and help companies show whether a treatment is working—information that can change a biotech firm's value much like a new product test result can affect a tech stock.

AI-generated analysis. How Rhea-AI works. Not financial advice.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

In a rare disease research dataset, SpliceAI2 identified 17% more disease-relevant variants compared to other splicing models

Illumina's genomic AI models now collectively enable researchers to identify 2x as many variants with predicted biological impact

SAN DIEGO, Oct. 8, 2026 /PRNewswire/ -- Illumina, Inc. (NASDAQ: ILMN) today introduced SpliceAI2, a new genomic AI model designed to address one of the central challenges in genomics: understanding the functional consequences of genetic variation. SpliceAI2 helps researchers identify disease-relevant splice variants that otherwise may go unnoticed. Predicting splice effects is key to resolving variants of uncertain significance in rare disease research, hereditary cancer testing research, and drug discovery. When applied to a rare disease research dataset, SpliceAI2 identified 17% more disease-relevant variants compared to other splicing models.

"Variant effect prediction tools, such as SpliceAI2, are among the key areas of focus for the BioInsight AI Lab," said Rami Mehio, senior vice president and general manager of BioInsight at Illumina. "As researchers work to elucidate the effect of mutations, we are uniquely positioned to unite genomic data and scientific expertise at scale, delivering the AI tools that can advance discovery and human health."

The BioInsight AI Lab focuses on three main areas of scientific impact: generating genomic and multiomic data to power AI tools; developing genomic AI models for variant effect prediction and prioritization; and training biological foundation models.

"Illumina is advancing AI to systematically shrink the portion of the genome that remains uninterpretable," said Kyle Farh, vice president of Illumina's BioInsight AI Lab. "Genomics has driven some of the most consequential genetic disease breakthroughs of the past two decades. Today, we are equipping researchers with the next generation of technology to help understand the underlying causes of disease."

SpliceAI2, PromoterAI, and PrimateAI-3D form a suite of genomic AI models designed to help researchers assess the functional impact of genetic variation across a broad spectrum of variant effect types, including splice, promoter, and missense variants. Together, these genomic AI models now collectively enable researchers to identify up to twice as many variants with predicted biological impact.

SpliceAI2 enables state-of-the-art variant interpretation across rare disease and population health research

Accurate interpretation of the functional consequences of genetic variation remains a central challenge for life science research. Splicing is the process by which RNA transcripts are assembled before protein production. Splice variants can disrupt transcripts, alter gene function, and interfere with protein function. Cryptic splice sites contribute to a significant burden of rare disease cases but are challenging to identify through traditional approaches. SpliceAI2 improves researchers' ability to identify and interpret these variants and provides accurate prediction of cell type-specific splicing.

A newly released preprint from Illumina's BioInsight AI Lab evaluated SpliceAI2's performance against the original SpliceAI and alternative splice prediction models. According to analyses of matched DNA and disease phenotype data from Genomics England cohorts, SpliceAI2 identified 17% more disease-relevant splice variants than other models. Analysis of matched whole genome and RNA sequencing data from the NIH's Genotype-Tissue Expression Portal showed that SpliceAI2 improved quantification of splice site usage by 34% compared to the next best model. SpliceAI2 results also aligned more closely with downstream protein effects seen in UK Biobank proteomic datasets and better captured the effects of splice variants across multiple population genomics datasets. 

Researchers worldwide already use genomic AI from Illumina to drive deeper insights into their sequencing data. The first generation of SpliceAI has become the splice variant prediction tool of choice for researchers, cited in over 3,400 publications to date. It is incorporated into guidelines from ClinGen, a clinical research body that sets standards for clinical genomics, for splice variant interpretation. Trained on a dataset that is 100 times larger than the original SpliceAI, SpliceAI2 advances the model through significantly improved prediction of splice variants, enabling the detection of more disease-relevant variants while reducing the overall number of variants requiring review.

Customers on the frontline of rare disease and drug discovery research can easily access SpliceAI2 through Illumina's BioInsight applications, including DRAGEN Annotation for AI-powered and traditional variant annotation, and Emedgene for germline research.

BioInsight is Illumina's data, software, informatics, and AI business, helping researchers, healthcare organizations, and biopharma companies transform genomic and multiomic data into biological insights. BioInsight enables customers to accelerate discovery, identify and validate drug targets, deepen understanding of disease biology, and advance precision medicine. BioInsight's growing portfolio includes the groundbreaking BioInsight AI Lab as well as resources such as the Billion Cell Atlas, which helps biopharma partners connect genetic signals to cellular mechanisms and biological function.

Learn more about SpliceAI2 here

Read the full preprint here

About Illumina

Illumina is improving human health by unlocking the power of the genome. Our focus on innovation has established us as a global leader in DNA sequencing and array-based technologies, serving customers in the research, clinical, and applied markets. Our products are used for applications in the life sciences, oncology, reproductive health, agriculture, and other emerging segments. To learn more, visit illumina.com and connect with us on X, Facebook, LinkedIn, Instagram, TikTok, and YouTube.

Contacts

Investors:
Illumina Investor Relations
858-291-6421
IR@illumina.com

Media:
Christine Douglass
PR@illumina.com 

View original content to download multimedia:https://www.prnewswire.com/news-releases/illumina-releases-spliceai2-to-help-advance-rare-disease-research-302902489.html

SOURCE Illumina, Inc.

FAQ

AI-generated questions and answers. How Rhea-AI works. Not financial advice.

How did Illumina's SpliceAI2 perform in rare disease research?

SpliceAI2 identified 17% more disease-relevant splice variants than other models in analyses of matched DNA and disease phenotype data from Genomics England cohorts. The evaluation appeared in a newly released preprint from Illumina's BioInsight AI Lab.

How can researchers access Illumina's SpliceAI2?

Researchers can access SpliceAI2 through Illumina's BioInsight applications, including DRAGEN Annotation for AI-powered and traditional variant annotation, and Emedgene for germline research.