Skip to main content

Data science integrations

Link JMP analyses and Jupyter or Colab notebooks to the Seal records they use, with the inputs and version needed to reproduce them.

Illustration of a seal beside a laptop showing a data table and an evidence binder linked by a red thread.

What these systems hold.

Notebooks and statistics tools hold analyses: the code or script, its inputs and parameters, and the outputs it produced, such as a regression, a capability study or a model fit.

FDA’s drug GMP regulations require laboratory records to include a record of all calculations performed in connection with a test, with units and conversion factors (21 CFR 211.194(a)(5)). When an analysis supports a result or a process decision, the organisation has to show which inputs and which version of the analysis produced it.

What Seal reads and writes back.

All three are set up with Seal’s API and Scripts. Seal reads exported .ipynb notebooks, or JMP analyses, graphs and reports, and links them to the records they describe, keeping the notebook version or the JMP data table and script, the inputs and the parameters with each result. Anything written back is agreed when the connection is scoped.

Seal
  1. JMPSource of truth agreed when scoped

    Reads Exported analyses, graphs and reports

    Writes back If agreed

  2. Google Colab, Jupyter NotebookSource of truth agreed when scoped

    Reads Exported .ipynb notebooks and their outputs

    Writes back If agreed

3 connections

Blueprints that use these records.

Start with one source.

Tell us what Neil needs to do and where the records live. We’ll scope the connection with your team.

Book a demo