Single cell datasets grow faster than most labs can quiz them. On November 2, 10x Genomics launched Sentira Single Cell, a cloud tool that lets researchers put biological questions to their Chromium data in ordinary language and assigns the planning to an AI agent.
The agent is Anthropic’s Claude. Given a hypothesis, it maps out an approach, writes out a plan, lets the researcher review and adjust it, carries out the steps and delivers findings. Everything lands in an editable notebook that preserves the code behind every figure.
Why the notebook matters more than the chat
Agentic tools in biology invite a fair question about verification. 10x’s answer: the reasoning trail ships with the output. Researchers can ask follow-ups, request changes, and inspect how a figure was reached. That keeps a human in the loop without making them write the pipeline by hand.
Hosted on 10x Cloud, the service supplies compute for Chromium datasets of any size and runs several analyses in parallel, so labs provision nothing themselves. Henry Long, scientific director at Dana-Farber Cancer Institute, noted that the questions themselves are old news; asking them in natural language is what changes how fast answers arrive.
A platform play, not a one-off feature
Sentira arrived as a computational platform in late September, plus a spatial analysis app in Early Access. A natural-language layer turns that infrastructure into something closer to a research collaborator. The company cites a standing bottleneck: labs produce rich, high-dimensional measurements, yet turning them into tested ideas remains slow.
Accessibility cuts both ways. Cheaper analysis widens who can query single cell data, and it raises the rigor bar, since a plausible figure nobody rebuilt by hand is easy to accept and hard to defend. Introductory credits are available now, then consumption pricing.
