Skip to main content
initializing...

    explore → analyze → predict → design → omics → reports

    chain tools.
    build workflows.
    keep the result.

    pick the model, then drive it from chat, curl, the CLI, or an MCP client. the tools chain into a pipeline. signed-in runs from the web app, the terminal, curl or MCP all write to the same hash-chained record. every result names the tool, parameters and run that produced it. you leave with what the run made — a structure, a ranked table, an export.

    how tools connect

    each tool's output becomes the next tool's input.

    no manual file management — Ivy handles the data flow across a full pipeline: sequence → structure → docking → ADMET / tox → synthesis route, passed forward.
    • protein structure

      protein structure → binding site detection

      structures scored for druggable binding pockets

    • docked complexes

      molecular docking → binding energy

      binding poses minimized and rescored with MM-GBSA energy estimates

    • validated compounds

      cheminformatics → ADMET prediction

      cleaned compounds ready for property profiling

    • active compounds

      database search → property prediction

      database hits profiled for drug-likeness

    • designed sequences

      antibody design → developability

      designed antibodies profiled for manufacturability

    • starting materials

      retrosynthesis → starting-material check

      routes end in starting materials checked against a purchasable-compound stock

    worked workflows

    one question can span several domains without you routing it.

    • intermediate · 5 steps

      binding site detection & characterization

      Identify and characterize druggable binding sites on protein targets using AI-powered pocket detection

      tool chain
      structure prediction → molecular docking + 3 more
      you leave with
      Ranked list of druggable binding sites with pocket properties, druggability scores, and docking-ready protein structures.
    • beginner · 5 steps

      compound library validation & QC

      Validate compound libraries for quality, diversity, and drug-likeness before screening campaigns

      tool chain
      3 tools
      you leave with
      Clean, validated compound library with quality metrics, diversity analysis, and flagged problematic structures.
    • beginner · 3 steps

      compound passport & export

      Dock and profile a lead compound, then package the runs into a compound passport and export it

      tool chain
      molecular docking → compound passport → export + 2 more
      you leave with
      A compound passport for the lead, plus an exported copy of it to share outside the platform.

    beyond the catalog

    no tool fits? the step still runs.

    a step the catalog doesn't cover (a custom filter, a plot, a one-off statistic) runs in the same chat, against what the tools themselves produced. it is recorded like every other step: signed-in runs leave a tamper-evident record you can check. with an account, you can export the whole session.

    the domains

    what each domain covers, and the plan that unlocks all of it.

    reference for when you already know what you're looking for. open a domain to see its example workflows.
    Workflow Domains
    Explore 6/7
    Analyze 5/9
    Predict 2/7
    Design 3/9
    Omics & Clinical 2/19
    Reports & Automation 2/9
    • Explore — All public sources
    • Analyze — CPU-based analysis
    • Predict — Basic folding and docking
    • Design — Fragment tools, cloning design, and antibody annotation
    • Omics & Clinical — Basic QC and CRISPR guide design
    • Reports & Automation — Basic passports and JSON/CSV/MD export

    every plan side by side, workflow by workflow: compare plans.

    ready when you are

    ready to build your first workflow?

    start with a template or chain tools your way.
    Loading...