Ahead of the 2026 meeting of the American Society of Human Genetics (ASHG), Illumina announced the launch of the BioInsight platform, a new unified software environment that provides tools to help scientists extract insights from multiomics datasets. According to the company, the platform provides scientists with seamless access to Illumina’s analysis applications, artificial intelligence tools, and curated data resources in a single environment.
The company plans to showcase the BioInsight Platform during the annual ASHG meeting being held in Montreal, Canada this year.
Illumina’s BioInsight business unit was created a year ago to address what the company perceived to be a gap in analysis capabilities for handling the scale and complexity of multiomics datasets. “With the BioInsight Platform, we are bringing customers one simple place to access the full breadth of our offerings across software, AI, and data,” said Rami Mehio, senior vice president and general manager of the BioInsight business. Moving forward, “customers that use the BioInsight platform will also have access to new capabilities in the future, empowering them to drive the next wave of biological insights for our industry.”
The list of tools available on the platform includes SpliceA12, a new genomic AI model that Illumina unveiled last week that is designed to identify disease-relevant splice variants. Being able to predict splice effects is key to resolving variants of uncertain significance in rare disease research, hereditary cancer testing research, and drug discovery. To date, its predecessor, SpliceA1, has been cited in over 3,400 publications and is incorporated into guidelines from ClinGen.
Details of SpliceA12’s framework, its training, and how it stacks up against other tools are provided in a pre-print published by scientists from Illumina, University of Oxford, University of California, San Francisco, and the New York Genome Center. According to the article, the tool “reconstructs complete transcripts directly from genomic sequence and considerably improves identification of splice-altering variants and quantification of their effects in diverse benchmarks spanning cryptic splicing, splicing quantitative trait loci, and massively parallel reporter assays.”
The data shared in the paper includes results from comparing SpliceA12’s to alternative splice prediction models on matched DNA and disease phenotype data from Genomics England cohorts. The findings show that SpliceA12 identified 17% more disease-relevant splice variants than other models. Furthermore, analysis of matched whole genome and RNA sequencing data from the National Institutes of Health’s Genotype-Tissue Expression portal showed that SpliceA12 improved quantification of splice site usage by 34% compared to the next best model.
Other findings indicate that SpliceA12’s results align more closely with downstream protein effects seen in the UK Biobank proteomic datasets. It also better captured the effects of splice variants across multiple population genomics datasets, according to data shared in the preprint.
Other applications included in the BioInsight platform include a new version of Illumina’s DRAGEN software solution. Version 4.6 expands where and how scientists can use the secondary analysis software and offers new deployment options including the DRAGEN software mode, which lets scientists run the tool across supported cloud and standard CPU environments.
Also available are Illumina’s BaseSpace and Connected Multiomics among other applications. Customers will also have access to the BioInsight AI lab, the research arm of BioInsight, which applies multiomic data and AI for tasks like variant prediction and training foundation models. The platform also features curated datasets for research spanning oncology, rare disease, population health, and multiomics.
Access is available via a credits-based system in lieu of a more traditional annual subscription. This gives scientists the flexibility to use the capabilities they need while freeing them to explore and adopt new technologies as needed.