py-lmd

The py-lmd package provides an opensource python interface to generate cutting directions for the Leica LMD6 & LMD7 laser microdissection microscopes. These cutting directions can be used to perform laser microdissection on arbitrary sample types.

py-lmd interfaces directly with your existing python workflows to make your samples accessible via LMD. For example you could use the output of your ML-based region detection on clinical tissue samples to export cutting shapes to excise these same regions on the LMD.

In addition, you can build reproducible workflows to calibrate slides, import SVG files and convert them to cutting masks, and can convert existing segmentation masks to cutting XMLs.

Installation

py-lmd has been tested with Python 3.8 and 3.9. To install the py-lmd library clone the Github repository and use pip to install the library in your current environment. It is recommended to use the library with a conda environment. Please make sure that the package is installed editable like described. Otherwise static glyph files might not be available.

We recommend installing the non-python dependencies with conda before installing py-lmd:

git clone https://github.com/MannLabs/py-lmd

conda create -n "py-lmd-env"
conda activate py-lmd-env
conda install python=3.9 scipy scikit-image>=0.19 numpy numba -c conda-forge
pip install -e .

If you are installing on an M1 apple silicon Mac you will need to install numba via conda instead of pip before proceeding with the installation of the py-lmd library.

conda install numba

Citing our Work

py-lmd was developed by Georg Wallmann, Sophia Mädler and Niklas Schmacke in the labs of Veit Hornung and Matthias Mann. If you use our code please cite the following manuscript:

SPARCS, a platform for genome-scale CRISPR screening for spatial cellular phenotypes Niklas Arndt Schmacke, Sophia Clara Maedler, Georg Wallmann, Andreas Metousis, Marleen Berouti, Hartmann Harz, Heinrich Leonhardt, Matthias Mann, Veit Hornung bioRxiv 2023.06.01.542416; doi: https://doi.org/10.1101/2023.06.01.542416

Contents:

• 1. Quick Start
  • 1.1. Installation from Github
  • 1.2. Generating Shapes
  • 1.3. Using the py-lmd tools
  • 1.4. Numbers and Letters
  • 1.5. Text
• 2. Using Segmented Images
  • 2.1. Background
  • 2.2. Different Coordinate Systems
  • 2.3. Getting started with the SegmentationLoader
  • 2.4. Overview of Configuration
  • 2.5. Processing Order
• 3. Example Notebooks
  • 3.1. Generate a Cutting XML from segmentation mask using the Segmentation Loader.
    • 3.1.1. Load Segmentation Mask
    • 3.1.2. Define cell sets
    • 3.1.3. Calibration points
    • 3.1.4. XML generation
    • 3.1.5. write to XML
  • 3.2. Generate cutting XML from exported results from QUPath
    • 3.2.1. Import libraries and define helper functions
    • 3.2.2. Import GEOjson regions
    • 3.2.3. Calibration points
    • 3.2.4. Clean up Dataframe
    • 3.2.5. Generate shape collection
    • 3.2.6. write to XML
  • 3.3. Convert SVG graphic into LMD excision instructions
    • 3.3.1. Visualize Results
    • 3.3.2. Export to XML
• 4. py-lmd Modules
  • 4.1. lmd.lib
    • 4.1.1. Collection
      • Collection
        • Collection.shapes
        • Collection.calibration_points
        • Collection.orientation_transform
        • Collection.add_shape()
        • Collection.join()
        • Collection.load()
        • Collection.new_shape()
        • Collection.plot()
        • Collection.save()
        • Collection.stats()
        • Collection.svg_to_lmd()
    • 4.1.2. Shape
      • Shape
        • Shape.from_xml()
        • Shape.to_xml()
    • 4.1.3. SegmentationLoader
      • SegmentationLoader
        • SegmentationLoader.check_cell_set_sanity()
        • SegmentationLoader.load_classes()
  • 4.2. lmd.tools
    • ellipse()
    • get_rotation_matrix()
    • glyph()
    • glyph_path()
    • makeCross()
    • rectangle()
    • text()