Over the last week, Tempus AI and the Mayo Clinic have each announced data platforms to link genomic data with longitudinal clinical data at scale: both targeting 1 million samples and both vying for a piece of the emerging market for AI-ready biological datasets.
On Sept. 11, Chicago-based Tempus said it will build a research platform containing 100,000 whole genomes linked to longitudinal clinical outcomes over the next several years, with a long-term goal of reaching 1 million genomes. The company said the effort marks the first deidentified multimodal whole-genome sequencing dataset built around disease populations and patient outcomes, expanding beyond the targeted gene panels the company has historically relied on. The new dataset will be integrated into Tempus’s existing data environment and will be accessible through Tempus Lens, the company’s platform for building and validating AI models.
Development of the dataset is already underway, and an initial version is available through Tempus’s Early Adopter Program; the company said it plans to onboard additional members in waves, with general availability planned for mid-2027.
“A large dataset is only valuable if you can turn it into insight,” Tempus founder and CEO Eric Lefkofsky said in a statement. “Tempus has spent the last decade building the infrastructure to connect diagnostics, multimodal clinical data and AI at scale. Adding whole-genome data linked to longitudinal outcomes makes that platform even more powerful and gives researchers a richer foundation to build AI models, generate new insights and ultimately improve patient care.”
On Sept. 15, Mayo announced it was forming Precure, a joint venture with Thermo Fisher Scientific that would also generate molecular data from 1 million biospecimens, linking genomic and proteomic information with longitudinal clinical data.
Mayo Clinic Forms Precure Joint Venture With Thermo Fisher to Study Early Disease Signals
Mayo Clinic has launched Precure, a new company formed with Thermo Fisher Scientific, which aims to better identify early biological signals of disease.
Separately, Tempus said it has been selected by the Advanced Research Projects Agency for Health (ARPA-H) to receive an award of up to $9.5 million to build an autonomous clinical AI agent for patients with heart failure. The funding will support both product development and clinical validation in a multi-center prospective study.
The proposed system will combine continuous surveillance of clinical data from electronic health records with consumer-generated data from heart rate monitors; the patient-facing AI agent would then manage symptoms, medications, rehabilitation and scheduling while feeding documentation back into the EHR.
“Heart disease remains the leading cause of preventable death in the United States, driving hundreds of billions of dollars in annual healthcare costs,” Brandon Fornwalt, senior VP of cardiology at Tempus, said in a statement. “In part, that’s because our healthcare system simply doesn’t have enough people to deliver the care we know these patients need. The ARPA-H ADVOCATE program is tackling this challenge head-on with technology, and we are thrilled to have been selected to build the first AI agent designed to operate as a fully autonomous member of the clinical team.”
The agent builds on Tempus’ existing AI products, including Tempus Next, a care-gap software the company said has screened more than 2.7 million cardiology patients.
Other Genomics News
EU Patent Court Grants Oxford Nanopore Injunction Against MGI Tech's CycloneSeq
The Munich Local Division of the European Union’s Unified Patent Court has granted Oxford Nanopore Technologies a preliminary injunction against sales of MGI Tech’s nanopore sequencers.
Veracyte Acquires Convergent Genomics for $150M, Adds Urine-Based Cancer Test
Veracyte has acquired Convergent Genomics, adding the company’s UroAmp platform and its proprietary urinary tumor DNA (utDNA) technology.
10x Genomics to Integrate Lunit's AI Pathology Tool for Oncology Biomarker Discovery
South Korea-based Lunit said 10x Genomics will incorporate the company’s AI-powered pathology platform, Lunit Scope IO, into its oncology biomarker discovery workflow.
Partillion Bioscience Wins $2.1 Million SBIR Grant to Scale Cell-Cell-Seq Platform
Partillion Bioscience said it has received a two-year, $2.1 million Small Business Innovation Research (SBIR) award from the National Institutes of Health’s National Institute of General Medical Sciences to expand its Cell-Cell-seq workflows.
Grail Sued for Patent Infringement by Zombie Australian Testing Company
An Australian company is suing Grail in US federal court, alleging patent infringement, even though it no longer performs the test that Grail’s Galleri test allegedly infringes.
GeneMind Biosciences, a China-based DNA sequencing instrument maker, said it has completed a Series D financing round of nearly 600 million yuan (approximately $89 million), according to a Sept. 14 report from Chinese business news outlet Jiemian News. The round follows a Series C raise completed in May 2025.
The round was co-led by Shanghai International Group Investment, Shenzhen Capital Group, Shengwei Rongquan Fund and Guoxin Fund, with participation from other investors, including several existing investors.
AstraZeneca disclosed that its Phase III SERENA-4 clinical trial for use of Etcamah (camizestrant) as a first-line treatment regardless of ERS1 mutation status did not meet its primary endpoint of progression-free survival.
“Whilst we are disappointed by the SERENA-4 outcome, it sharpens our focus on maximising the number of patients who can benefit from Etcamah today based on SERENA-6 and reinforces the importance of ESR1 testing for patients on first-line therapy,” Susan Galbraith, executive VP of oncology hematology research and development at AstraZeneca, said in a statement.
The trial miss is “ultimately good” for Guardant Health, stock analyst Subbu Nambi of Guggenheim Securities wrote in a Sept. 14 note to investors, as Guardant360 remains the required test to screen for ESR1 mutations before patients can qualify for Etcamah.
Earlier this month, the FDA approved the Guardant liquid biopsy test as a longitudinal companion diagnostic to identify emerging ESR1 mutations in patients on first-line therapy, which could precipitate a switch to treatment with Etcamah.
Mission Bio has launched Tapestri Triomics, a new expansion of its single-cell platform that simultaneously analyzes DNA, RNA, and protein in the same cell.
Researchers led by Terra Lasho of the Mayo Clinic have already applied the new product to bone marrow from patients with ASXL1-mutant myeloid disease to compare mutant and wild-type clones and how they may contribute to disease progression.
“Triomics allows us to move beyond identifying which cells carry an ASXL1 mutation and begin asking what that mutation means for their biology,” Lasho said in a statement, providing a “deeper understanding of myeloid disease progression.”
My Omics Reading List
Miao, J., Davis, J.R., Zhang, Y. et al. “Reimagining research papers as interactive and reliable AI agents.” Nature. Sept. 16, 2026.
A team of researchers at Stanford University reimagines the peer-reviewed paper as a dynamic document powered by AI agents.
The Paper2Agent framework “ [converts] a paper into an AI agent that functions as a virtual corresponding author, exposing its manuscript, supplementary materials, datasets, code and workflows as active, agent-native knowledge rather than static text.”
–
Fu et al. “Cell-type-resolved somatic variant discovery from bulk long-read sequencing.” medRxiv. Sept. 4, 2026.
An international team led by Fritz Sedlazeck of Baylor College of Medicine present a preprint on SniffCell, a computational tool that uses methylation data to enable cell type-specific variant calling from bulk long read samples.
“SniffCell turns native bulk long-read WGS into a cell-type-aware mutation discovery experiment,” they wrote.
“This allows much deeper insights into tandem repeats and other complex regions with cell type specificity on a population scale,” Sedlazeck told Ion Genomics in an email.
What I’m reading
After finishing The Odyssey for my nighttime reading I’m going a little bit more modern for my next choice: Aleksandr Solzhenitsyn’s Cancer Ward.