BioTuring and Human Cell Atlas Collaborate to Accelerate Data Exploration Using Next-Generation GPU Computing

San Diego, CA & Boston, MA 

The Human Cell Atlas (HCA) is a global, open consortium founded in 2016 with over 4,000 members across more than 100 countries, working to map every cell type in the human body and create a 3D Atlas to transform understanding of biology and disease. Organized through 18 Biological Networks (e.g., lung, heart, liver, immune system) and including four Regional Networks (Asia, Latin America, and Middle East), the initiative is creating a comprehensive reference map of human cells, to advance disease diagnosis and treatment worldwide.

Through BioTuring’s partnership with HCA, that vision advances to a new stage. HCA Atlases can now be explored in Talk2Data, an advanced GPU-accelerated platform built on NVIDIA accelerated computing, that is tailored for single-cell RNA, spatial, and other multiomics data. Researchers can perform sophisticated analyses including pseudotime trajectory inference, differential gene expression analysis, and high-dimensional clustering, and many other advanced analytics and visualization with enhanced interactivity – all in real-time, thanks to the GPU-accelerated algorithm suits, developed by the BioTuring team.

Son Pham, CEO of BioTuring, said ‘The Human Cell Atlas is creating a trusted reference of human cells. By bringing HCA data into Talk2Data, we ensure scientists can access and analyze this resource more easily, enabling faster progress in understanding disease and developing new treatments.’

The collaboration enables scientists to execute cross-atlas differential analyses, compute correlation metrics among cell populations, and validate marker gene expression profiles in any defined cell populations in real time, among other data science workflows. Such capabilities empower researchers to dissect cellular heterogeneity at a resolution spanning thousands of distinct combinations of cell types, cell states, conditions, treatment, tissues, and sampling-sites, rather than traditional broad cell type categorizations. This is crucial for elucidating molecular mechanisms underpinning health and disease states.

Furthermore, by anchoring disease sample analyses against a robust healthy human cellular reference map, this platform enables deeper mechanistic understanding which is vital for biomarker identification, drug target discovery, and translational research pipelines.

Besides the traditional graphical user interface, analysis using natural language makes the data even more accessible. Through Model Context Protocol (MCP), the platform can also be accessed by other AI agents. This enables prompt-based bioinformatics workflows and agent-to-agent collaboration – a step toward embedding the Human Cell Atlas into both human-led and autonomous scientific exploration.

John Randell, HCA Executive Director, said: ‘HCA is pleased to be collaborating with BioTuring so that people can now explore HCA Atlases with Talk2Data.  Many of the HCA Biological Networks have already released their initial Atlases with the rest to follow over the next year. These high-quality, community-generated reference Atlases will advance our understanding of human biology and accelerate the development of new foundation models that researchers and clinicians can use to understand disease.’  

Acceleration for the Global Research Community

The scale of HCA’s data makes interactive exploration technically demanding. By leveraging NVIDIA DGX B200 systems, built on the NVIDIA Blackwell architecture, BioTuring can address this challenge by compressing and processing single-cell and multi-omics data directly on the GPU – allowing users to analyze hundreds of millions of curated single-cell data in real time by simply asking complex biological questions in plain English. High-throughput bit-shuffling algorithms remove costly data transfers and cut computation from hours to seconds, enabling real-time exploration of billion-scale datasets that span thousands of variables.

For the HCA community, this means scientists can move from a question to a result – and then to the next question – without the delays that can interrupt the flow of discovery.

Explore integrated HCA datasets today via BioTuring Talk2Data: http://talk2data.bioturing.com and the HCA Data Portal: data.humancellatlas.org

About BioTuring

BioTuring is a GPU-powered bioinformatics company that delivers ultra-fast, AI-driven analytics for single-cell and spatial omics. With the world’s largest collection of curated public studies and deep-learning algorithms that accelerate discovery up to 1,000×, BioTuring enables life-science innovators to turn data into breakthroughs.

About Human Cell Atlas

The Human Cell Atlas (HCA) is an international collaborative consortium whose mission is to create comprehensive reference maps of all human cells—the fundamental units of life—as a basis for understanding human health and for diagnosing, monitoring, and treating disease. The global HCA community is producing high quality Atlases of tissues, organs and systems, to create a milestone Atlas of the human body to benefit humanity across the world.  https://www.humancellatlas.org