CasTuner - Python Implementation

CasTuner Logo

Deploy MkDocs to GitHub Pages Create Release License: MIT Python Version
DOI

Quantitative modelling of analog gene tuning

CasTuner combines a degron-controlled CRISPR/Cas repressor with single-cell flow cytometry to tune endogenous gene expression in an analog way rather than as a simple ON/OFF switch.

This repository contains an independent Python implementation of the CasTuner analysis framework. It reproduces the original R-based pipeline for:

  • kinetic fitting of degron–repressor dynamics,
  • dose–response (Hill) curve fitting,
  • ODE-based modelling of repression and derepression,
  • single-cell noise analysis,
  • model-driven design-space exploration,
  • uncertainty and sensitivity analysis, and
  • automated reporting.

The focus of this documentation is not only how to run the code, but why each step exists and what biological question it answers.

What problem does this solve?

Many processes in development and cell fate decisions are dose-dependent: small changes in the amount of a transcription factor can flip decisions such as pluripotency vs differentiation or dosage compensation. CasTuner was designed to measure and control such processes by:

  • tuning Cas-derived repressors with a ligand-controlled degron,
  • reading out endogenous reporters (e.g. Esrrb-mCherry, Nanog-mCherry, STAG2-EGFP),
  • quantifying dynamics, dose–response and noise at single-cell resolution.

This Python port turns that biological framework into a reproducible, Snakemake-driven workflow that can be extended to new genes, constructs or model variants.

How the documentation is organised

  • Biological Problem – the conceptual and experimental context.
  • Data & Experiments – what was measured and how it is organised.
  • Analysis Pipeline – step-by-step explanation of each script and model:
    • Step 1 – Kinetic Fits (up/down half-times)
    • Step 2 – Dose–Response (Hill curves)
    • Step 3 – ODE Repression/Derepression
    • Step 4 – Single-cell Noise
    • Step 5 – Design Space Scan
    • Step 6 – Goodness of Fit
    • Step 7 – Design Selection
    • Step 8 – Automated Report
    • Step 9 – Sensitivity & Uncertainty
  • API Reference – automatically generated from the codebase.
  • Reproducibility & Code – environments, Snakemake, validation vs the original R workflow.
  • References – primary CasTuner paper and related work.

For a quick practical overview (installation, Snakemake usage, validation tables), see the project-level README.md.

Launch Binder

Note: It may take a few minutes to launch the environment on Binder.

  1. Notebook: Quickstart with Toy Dataset

Binder

  1. Notebook: ODE Simulation with Toy Dataset

Binder

  1. Notebook: Real FCS data analysis

Binder