The Foundation of What Comes Next: BioTuring leverages NVIDIA Blackwell GPUs to power multi-omicsat a new scale for scientists and AI agents

SAN DIEGO, CALIFORNIA – BioTuring has announced the launch of the new Talk2Data platform – a GPU accelerated multi-omics database built on NVIDIA Blackwell accelerated computing – enabling real-time analysis of over 500 million high-quality, expertly curated single cells across multiple omics layers. The platform offers an intuitive interface for human scientists to explore large-scale data interactively, and integrates seamlessly with virtual scientists (AI agents) via Model Context Protocol (MCP).

Breaking Computational Limits

Modern biology is generating data faster than it can be analyzed. The Human Cell Atlas has mapped 100M+ cells, and the Billion Cells Project aims for 1 billion to advance AI in biology. Spatial transcriptomics pushes scale further – one slide alone can hold 500M precisely mapped transcripts, and thousands of slides yield unprecedented spatial data.

This growth in biological data parallels BioTuring’s own trajectory – with our multi-omics database now poised to reach 1 billion cells by year-end. Asking key questions on these massive datasets – such as finding co-expressed genes, identifying cell types and contexts with highest expression, discovering markers, mapping colocalization, studying cell-cell communication, and defining tissue niches – can require terabytes of processing.

The NVIDIA DGX B200 , built on the NVIDIA Blackwell architecture, is revolutionizing single-cell and multi-omics data analysis by enabling real-time processing and analysis of massive datasets. This system delivers 40x faster analysis for the Talk2Data platform than previous solutions and can scale to over a billion cells. BioTuring scientists leverage this power to perform high-throughput, bit-shuffling algorithms directly on the GPU, which is the key to unlocking its performance for cellular data processing. This method eliminates costly data transfers and cuts computation time from hours to seconds.

Changing How Scientists Work

With computational limits removed, scientists can start with a broad question, see results in seconds, and branch into new comparisons or deeper analyses within the same session. Talk2Data includes a natural language interface that allows human scientists to query and analyze multi-omics data conversationally. Through MCP, the platform can also be accessed by other AI agents – enabling automated, agent-to-agent workflows where virtual scientists trigger analyses, integrate results into larger pipelines, or collaborate with other computational tools. This dual accessibility makes large-scale, real-time analysis a seamless part of both human-led research and autonomous AI-driven discovery.

Advancing Science Together

The release of Talk2Data New Version reflects a shared commitment to give scientists the ability to work at the speed of their ideas, without being held back by infrastructure limits.

By combining BioTuring’s expertise in multi-omics curation and analysis with NVIDIA in accelerated computing, the collaboration delivers not just faster results, but a more fluid, interactive way to work with data.

The Foundation of What Comes Next

With Talk2Data on NVIDIA DGX B200, BioTuring is setting the stage for a new era where both human and virtual scientists can work at the speed of thought. The discoveries that will define the next decade in biology will come from those – human or AI agents – who can ask more questions, test more ideas, and see more connections – turning complexity into discovery.

Explore the new Talk2Data and see how your research can move at the speed of curiosity:
http://talk2data.bioturing.com/

About BioTuring

BioTuring provides GPU-accelerated solutions that enable researchers to process, visualize, and interpret complex biological data with unprecedented speed up to x1000 and ease. BioTuring is committed to democratizing access to cutting-edge bioinformatics, accelerating discovery in life sciences.