TLDR: TileDB, Kepler AI, and Tahoe Therapeutics have announced a strategic partnership to launch a pioneering AI-powered platform designed to revolutionize single-cell data analysis. This collaboration integrates TileDB’s multimodal database technology, Kepler AI’s advanced agent systems, and Tahoe Therapeutics’ extensive single-cell datasets, including the 100-million-cell Tahoe-100M Atlas, to accelerate drug discovery and computational biology.
SAN FRANCISCO, September 30, 2025 – A significant leap forward in computational biology and drug discovery has been announced with the formation of a groundbreaking partnership between TileDB, Inc., Kepler AI, and Tahoe Therapeutics. The three entities are combining their distinct expertise to create the first public-facing platform that will empower researchers to conduct sophisticated AI agent-based queries on massive single-cell datasets at an unprecedented scale. This collaboration is poised to transform how scientists analyze complex biological data, ultimately accelerating the identification of new therapeutic pathways.
The core of this innovative platform lies in the synergistic integration of each partner’s strengths. TileDB brings its robust multimodal database technology, uniquely capable of handling the multidimensional nature of single-cell genomics data while maintaining the high performance required for real-time AI agent queries. This capability is crucial for managing the vast and intricate datasets inherent in single-cell analysis. Stavros Papadopoulos, CEO of TileDB, emphasized the significance of their technology, stating, “TileDB is the only database technology capable of handling the multidimensional nature of single-cell genomics data while maintaining the performance necessary for real-time AI agent queries. This partnership represents exactly the kind of transformative application we envisioned when we built our multimodal database platform.”
Kepler AI contributes its advanced AI agent systems, which will enable researchers to interact directly with full-resolution single-cell data. Previously, the sheer scale of datasets like Tahoe-100M often constrained AI agents to working with pseudo-bulked data, offering only a limited view of cellular complexity. With this partnership, Kepler AI’s agents can now perform real-time queries across multidimensional single-cell datasets, providing scalable analysis from individual cells to population-level insights. This allows for the discovery of biomarkers and therapeutic pathways that were previously obscured in less granular data.
Tahoe Therapeutics provides the critical biological data, including its open-sourced Tahoe-100M Atlas. This giga-scale single-cell perturbation atlas comprises over 100 million transcriptomic profiles derived from 50 cancer cell lines exposed to 1,200 small-molecule perturbations. The Tahoe-100M dataset, generated using Tahoe Therapeutics’ Mosaic technology, was previously challenging to query in standard formats due to its immense scale and complexity. By leveraging TileDB’s architecture, the atlas is now available in an interoperable, highly scalable cloud-native platform, allowing pharma and biotech teams to utilize it in conjunction with their own datasets for training foundational models or running real-time queries.
The joint product, dubbed TahoeDive, integrates Tahoe’s landmark Tahoe-100M atlas directly into Kepler AI’s enterprise-grade bioinformatics AI agent platform. This allows scientists to ask and answer complex questions using natural language, eliminating the need for extensive coding. Key capabilities of TahoeDive include natural-language queries over 100 million single-cell transcriptomes, automated code generation, analysis, and visualization, context-aware reasoning, and scalable compute with secure data access. Early tests by Tahoe’s scientists demonstrated that discoveries that previously took days of bespoke analysis could be reproduced in minutes using TahoeDive.
Also Read:
- Saama Unveils Advanced Modular AI Agents to Revolutionize Clinical Development
- GoodData Unveils Next-Generation Full-Stack Data Intelligence Platform with Advanced AI Integration
This collaboration marks a significant milestone in the application of artificial intelligence to biological research, promising to accelerate drug discovery and deepen our understanding of cellular mechanisms by making vast, complex single-cell datasets more accessible and actionable for the scientific community.
