10x Genomics Launches Sentira Single Cell Powered by Anthropic Claude

Single-cell biology generates massive datasets, but analyzing that data remains an exhausting hurdle. Biologists often spend weeks writing custom code to cluster cells and interpret RNA expression. Now, 10x Genomics is handing that burden to autonomous AI agents.

The life sciences hardware leader launched Sentira Single Cell, a cloud analysis tool powered by Anthropic’s Claude models. Built for users of its flagship Chromium platform, Sentira allows biologists to explore complex genetic data using simple text prompts.

The system tackles the field’s chronic last-mile problem. Rather than waiting for bioinformatics teams to write scripts, researchers can chat directly with their experiments on 10x Cloud.

The Last-Mile Problem: Why Single-Cell Analysis Stalled in Code

Single-cell RNA sequencing isolates thousands of individual cells to measure gene activity. The technology revolutionized cancer biology, immunology, and neuroscience. Yet analyzing the resulting data required advanced programming skills in R or Python.

Every research question is unique. Because off-the-shelf software could not handle every custom workflow, wet-lab biologists faced long queues waiting for bioinformaticians to build custom scripts. 10x Genomics Chief Technology Officer Michael Schnall-Levin described this as a long-standing roadblock.

“Ask any single cell researcher for their biggest experimental bottlenecks, and they will say data analysis. Going back to the beginning of the field, there has always been a last mile problem. What might have taken someone weeks before could take hours.”

Michael Schnall-Levin, PhD, Chief Technology Officer at 10x Genomics

With Sentira, the company hopes to democratize data discovery. Researchers without coding skills can question their datasets directly, spotting cellular patterns that manual reviews overlook.

From Stanford’s Virtual Lab to 10x Cloud: How Sentira Works

The architecture behind Sentira grew out of academic research. In early 2026, 10x Genomics collaborated with Dr. James Zou, Associate Professor of Biomedical Data Science at Stanford University. Zou pioneered the Virtual Lab framework, which was first demonstrated in a 2025 Nature paper on AI-designed SARS-CoV-2 nanobodies.

Sentira operates on 10x Cloud, supplying managed computing power without server configuration headaches. The platform features an agentic workflow:

• Natural Language Dialogue: Biologists type biological questions in everyday English, asking the system to find differentially expressed genes or identify rare immune subtypes.
• Autonomous Scripting: In the background, Claude generates and runs Python and R analysis code specifically tailored to Chromium data structures.
• Parallel Execution: The cloud engine spins up compute instances to run multiple clustering tests and statistical verifications at the same time.
• Strict Data Privacy: 10x Genomics monitors system telemetry to improve software responsiveness, but confirmed that customer genomic data never flows back into model training.

Early adopters note that foundation model tools are advancing fast. Dr. Henry Long, Scientific Director at the Dana-Farber Cancer Institute, noted that 10x must continue updating Sentira to match general frontier tools like Anthropic’s Claude Science and OpenAI’s Rosalind.

The Peril of Silent Mistakes: Why Scientists Urge Caution

Despite the massive productivity leap, leading scientists urge careful oversight. The primary danger with automated bioinformatics is that machine-generated results rarely look broken.

Dr. Luciano Martelotto, Director of Global Market Development for Single Cell at Waters Biosciences, warned about the psychological trap of clean outputs. AI software does exactly what researchers ask, but it lacks the scientific skepticism that spots false positives.

“Nothing in an AI-generated analysis looks broken. The code runs, the plots are pretty, the story makes sense. The mistakes that matter are the silent ones. What it doesn’t do is get that uneasy feeling when a result looks too good. That feeling comes from having watched data go wrong, and I don’t think it can be prompted or coded into existence, not yet, at least.”

Dr. Luciano Martelotto, Single Cell Market Development Director at Waters Biosciences

Dr. Holger Heyn of the Centro Nacional de Análisis Genómico echoed this concern. He pointed out that while AI can cluster millions of cells, models still struggle with the genuine biological variation found in living human tissue.

Factsheet: Comparing Single-Cell Bioinformatics Workflows

The following table outlines how Sentira Single Cell compares to manual scripting and generalized language models:

Pricing and Packaging: 10x Cloud Credits Structure

Sentira uses a consumption-based pricing model powered by 10x Cloud Credits. A standard exploration session costs roughly 12 credits. Reaching final experimental conclusions typically requires around 60 credits.

To encourage trial, 10x Genomics grants 100 free credits with every user’s first uploaded Chromium dataset. Ongoing credit packs feature tiered volume discounts:

• 100 Credit Pack: $120 ($1.20 per credit)
• 1,000 Credit Pack: $1,150 ($1.15 per credit)
• 2,000 Credit Pack: $2,200 ($1.10 per credit)
• 10,000 Credit Pack: $10,500 ($1.05 per credit)
• 25,000 Credit Pack: $25,000 ($1.00 per credit)

This launch matches a broader industry pivot toward single-cell automation. Earlier today, UCL spinout Octopath AI raised £1.5M for cancer biopsy pathology using single-cell foundation models. Across genomics and pathology, artificial intelligence is transforming raw biological samples into rapid clinical answers.

Automated agents are also expanding into corporate treasuries. Read our analysis on how Aion launched an autonomous AI agent across business banking and payments to handle bill payments, credit lines, and corporate banking under strict human controls.

Frequently Asked Questions