AI Blocks › Build and ship software
Single-cell deep learning analyst
Give it single-cell data. Get your datasets merged, batch effects fixed and cell types found with scvi-tools.
Single-cell deep learning analyst is a free AI skill for the job "Build and ship software". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · Apache 2.0 licence · Made by Anthropic · Level: expert · Thinking: high
Made by Anthropic. See the original on GitHub; we copied it as it is, with its licence.
What it does
Single-cell datasets merged, batch effects removed and cell types found with scvi-tools, the deep learning toolkit for single-cell data. Your samples then line up so real biology shows through. It starts with a model table: scVI for plain RNA integration, scANVI when you have labels, totalVI for RNA plus protein, PeakVI and MultiVI for chromatin data, and DestVI for spatial data. It points you to the matching step-by-step reference, ready scripts and notes on GPU setup and fixing errors. Tasks include batch correction, label transfer to a reference, joint analysis of several data types and a latent map of cells. You need Python, scvi-tools and ideally a GPU.
You give it
- Single-cell data, often an AnnData file
- Batch or sample information
- Labels, if you have them (optional)
- Python with scvi-tools
- Codebase (optional)
You get
- The right model for your data type
- Integrated data with batch effects reduced
- Transferred cell type labels
- A latent representation for downstream work
Use it when
- You have samples from different labs and batch effects hide the signal.
- You want to label new cells using a reference dataset.
- You work with CITE-seq or multiome data.
- You need a latent space for clustering or comparison.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run code.