FigureYa Help

Why We Now Recommend Wisp Science

Update (2025): This page originally recommended VS Code because its AI chat interface (e.g., Copilot Chat) lets beginners ask AI while coding. A year later, >Wisp Science takes this idea to its logical conclusion — it's not just an editor with an AI sidebar, but an AI research workbench built specifically for scientific computing. We now recommend Wisp Science for running FigureYa.

From VS Code to Wisp Science

The original case for VS Code over RStudio was: AI assistance means beginners are no longer stuck on syntax and configuration. Wisp Science goes further:

Original reason to choose VS Code How Wisp Science evolves it
Integrated chat panel AI is native — not just chat, but an agent that reads/writes files, runs code, and manages projects
Multi-language support (R, Python, etc.) Built-in persistent Python and R kernels; variables persist across cells and turns
Required manual extension installs & config Works out of the box: bundled MCP bioinformatics tools (PubMed, GEO, ~80 databases), SSH/WSL/GPU remote compute
AI could only answer questions AI can execute full analysis: retrieve data → write code → generate figures → record decisions

Key Advantages of Wisp Science

1. AI-Native Research Workbench

  • Bring your own model: Supports OpenAI-compatible and Anthropic models, or drive Codex / Claude Code over ACP.
  • Beyond chat: The agent reads and writes project files, runs shell, loads reusable Skills (SKILL.md). Approval gates stay on by default.
  • Project memory: Restart and the full history is back. Undo a turn's file edits. Use @ for files, # for saved sessions, / for skills.

2. Persistent Python/R Kernels

  • Each conversation gets its own isolated kernel; variables persist across cells and turns.
  • Parallel conversations never share state.

3. Remote Compute

  • Register local, WSL, and SSH hosts once.
  • Probe hardware, submit long Runs with live logs.

4. Built for Science

  • ~80 scientific databases (PubMed, GEO, etc.) via bundled MCP servers.
  • Offline previews for notebooks, PDFs, Office files, and images.
  • Isolated explorations to try a direction without touching the mainline.
  • Publication Workspace that freezes manuscript revisions into verifiable Evidence Capsules.

5. Local-First, Your Data Stays Yours

  • Your data, conversations, and credentials stay on your machines.
  • No background syncing; encrypted manual sync and project transfer under your control.

Download & Install

Visit the >Wisp Science download page and choose your OS and processor:

Platform Package
Windows Signed MSI / NSIS
macOS Signed, notarized .dmg (Apple Silicon + Intel)
Linux .deb / AppImage (x86_64 + aarch64)

After installing, open the bundled demo (no API key needed) to see a full RNA-seq trajectory. Then add a model in Settings → Models and start a project.

Comparison: RStudio vs VS Code vs Wisp Science

Feature RStudio VS Code + AI plugin Wisp Science
Default R support Yes Requires extension Built-in R kernel
Default Python support Limited Requires extension Built-in Python kernel
AI assistance None Chat panel Native agent: reads/writes files, runs code
Multi-language R, Python Multi-language R, Python + MCP science databases
Persistent kernels R only None Python and R persist across turns
Remote compute None None SSH / WSL / GPU
Project memory None None Full history, undoable
Database integration None None ~80 scientific databases

Next Steps

Note: If you are still using VS Code, the original guide remains available as a reference, but we strongly recommend upgrading to Wisp Science for the full AI-powered research experience.