STELA -the world’s largest clinically linked spatial biology atlas
Spatial Tissue Embedding Learning Atlas (STELA) is a multinational spatial data generation initiative powered by a collaboration with 10x Genomics, Inc. (Nasdaq: TXG) and Broad Clinical Labs.Bioptimus, a global AI company announced the launch of STELA on March 25, 2026.
Bioptimus is a global AI biotech business that developed the world’s first universal foundation model for biology. By integrating cutting-edge AI with vast, multimodal, proprietary data production, Bioptimus is creating a unifying framework that connects all scales of biology, from molecules to patients, and delivers interpretable, dynamic, and actionable insights. Bioptimus models are used by 16 of the top 20 pharmaceutical companies.
While foundation models for language have thrived on massive digital datasets, biology has long lacked the standardised, high-quality data scale required for a similar breakthrough, particularly for clinical data. Bioptimus is closing this gap by constructing the data infrastructure required to power M-Optimus, the first multimodal and multiscale world model of biology. STELA will act as M-Optimus’ data backbone, producing huge datasets aimed at decrypting the complicated architecture of human tissues. M-Optimus will use this massive multimodal repository to map how molecular and cellular interactions drive disease in fields such as oncology and inflammation, eventually allowing researchers to predict patient responses to novel therapies, accelerate drug development, and design more effective immunotherapies.
Starting with 10x Genomics’ Xenium spatial transcriptomics and designed to integrate additional spatial and molecular profiling technologies over time, STELA will generate harmonised datasets integrating high-resolution spatial transcriptomics, matched histopathology imaging, multi-omics data (e.g., genomics, transcriptomics, and proteomics), and longitudinal clinical records. The effort intends to profile up to 100,000 patient tissue specimens from three continents: the United States, Europe, and Asia, creating the world’s biggest clinically linked, spatially profiled, multimodal patient data atlas.
Participating hospitals and research institutes will donate samples following standardised protocols in exchange for access to advanced spatial characterisation and foundation model capabilities. This cooperation enables clinicians to transform raw data into meaningful insights by using more precise diagnostic and treatment procedures. STELA creates a unified platform for data collection, processing, storage, and AI model building on a global scale, laying the groundwork for biological AI’s future.
Today, the majority of patients’ diagnostic data is utilised to make decisions that exclusively affect them. “We envision a world in which every patient can contribute insights to better inform future patients’ care and treatment outcomes, just as patients with other diseases, heritage, and even from the past do today”, said Jean Philippe Vert, PhD, Co-Founder and CEO of Bioptimus. “STELA is the fuel to power M-Optimus, allowing us to map the intricate interactions between cells and tissues, across indications, at unprecedented scales, unlocking a new era of precision medicine.”
Partnerships that Shape Industry
Bioptimus and 10x Genomics’ partnership unites cutting-edge spatial biology technology with extensive AI foundation model development, combining complementary skills to transform the creation and use of biomedical data. STELA will provide highly standardised spatial datasets using the Xenium platform, enabling the large-scale, repeatable data collection needed for large-scale AI model development across collaborating institutions globally.
According to Serge Saxonov, CEO and co-founder of 10x Genomics, “many of the most significant questions in medicine come down to understanding how cells interact within complex human tissues.By enabling spatial profiling at unprecedented scale, STELA will generate foundational datasets that allow researchers to connect the underlying biology with disease outcomes, unlocking new insights that can accelerate and improve therapeutic discovery and development.”
Through a historic partnership with the Broad Clinical Laboratories, Bioptimus will anchor its STELA program, connecting cutting-edge AI research with industrial-scale data creation. This collaboration creates a vast reservoir of high-resolution spatial transcriptomics data by processing biological samples at scale using Broad’s high-throughput spatial biology capabilities. In addition to producing data, the two companies will collaborate to create prediction tools and next-generation AI-driven quality control measures that will maximise test performance and automate biological insights. This partnership guarantees that the STELA effort is based on a foundation of extraordinary technical precision, expediting the creation of revolutionary AI models for the life sciences by fusing Bioptimus’s proprietary AI models with Broad’s industry-leading laboratory operations.
“To unlock the true clinical potential of spatial biology, we must pair massive-scale data generation with uncompromising data quality,” Niall Lennon, Chief Scientific Officer of Broad Clinical Labs added. “By combining our high-throughput laboratory workflows with Bioptimus’s advanced AI, we are co-developing next-generation quality control metrics that ensure the highest data integrity. This unprecedented technical precision guarantees that the insights generated by STELA can be confidently translated into actionable clinical diagnostics and precision therapies.”
Picture Source: Bioptimus