Snakemake workflow: 3DChromTrans

Measure distances between two types of markers during FISH Assay after Chromosomal Translocation event

Author

• Raul Gomez Riera

If you use this workflow in a paper, don't forget to give credits to the author by citing the URL of this repo.

Image acquisition settings for data generation

The Fluorescence In Situ Hybridization (FISH) assay was carried out with the following fluorophores:

• Texas Red, Emission Wavelength=614nm
• Alexa Fluor 488, Emission Wavelength=517nm

For nuclei detection, cells were stained with

• DAPI, Emission Wavelength=465nm

3D multiplex images of whole cells were acquired with ZEISS LMS 980 microscope in Airyscan mode.

• Objective Immersion="Oil" LensNA="1.4"
• Model="Plan-Apochromat 63x/1.4 Oil DIC (UV) VIS-IR M27"
• NominalMagnification="63"
• WorkingDistance="193.0" WorkingDistanceUnit="um"
• Zoom="3.6"
• Voxel Size: 0.073x0.073x0.130 um

NOTE : Lateral resolution of 0.073 um must be achived

Installation

You will need a current version of snakemake on Linux OS to run this workflow. To get snakemake please follow the install instructions on their website, but in brief once conda and mamba are installed you can install snakemake with:

mamba create -n snakemake -c conda-forge -c bioconda snakemake

Afterwards you can activate the conda environment and download the repository. And all additional dependencies will be handled by snakemake.

conda activate snakemake
git clone https://github.com/rgomez-AI/3DChromTrans.git

Create required environments by going to the directory 3DChromTrans/workflow

where Snakefile is located and execute the following command:

snakemake --cores all --use-conda --conda-create-envs-only Data_Analysis

Workflow Diagram

Component	Script	Description
split_channels	DVsplit_headless.ijm	Convert .DV file into .TIF file per channel
reslice_scale	ResliceZandScale_headless.ijm	Make isotropic the 3D space by image interpolation
cellpose	cellpose	Detect and create nuclei mask
label_conversion	LabelConversions_headless.ijm	Convert nuclei mask into labeled image
cellprofiler	3D_Distance_LowResolution.cppipe	Detect and provide the coordinates per nucleus of both markers
R	DataAnalysis_headless.R	Calculate all possible distance combinations between markers
DataAnalysis	labeled2Dist_headless.ijm	Generate 3D distance maps for visual inspection of the results

Input

Acquired images should be chromatic corrected before going to workflow execution.

This can be done by installing and executing Chromagnon software instructions on their website

and set the output as delta vision file format (.dv)

Sample dataset provided by Anna Oncins is an image of Mantle cell lymphoma: download

Running

To execute change current directory to the directory workflow where Snakefile is located.

snakemake --cores all --use-conda Data_Analysis

Output

As an output there are two files results/Results_in_um_Nuclei.xlsx which contain the following information about each Nucleus.

• Location_Center_X: Absolute location in X coordinate.
• Location_Center_Y: Absolute location in Y coordinate.
• Location_Center_Z: Absolute location in Z coordinate.
• EquivalentDiameter: According to the measured volume the expected diameter of the corresponding sphere.

All distances combination between the two markers per each nuclei and per image are also provided

• Min_Dist: Minimum distance value coming from all combination of distances between markers.

And results/Results_in_um_Markers.xlsx file with the following information about each marker:

• Location_Center_X: Absolute location in X coordinate.
• Location_Center_Y: Absolute location in Y coordinate.
• Location_Center_Z: Absolute location in Z coordinate.
• Min_Dist2Suf: Minimal distance to the nucleus surface.
• Norm_Dist: Is the Min_Dist2Suf divided by minimal distance from the nuclear center to the suface.

Report generation

For report generation snakemake required pygments module and it can be installed with:

pip install pygments

Afterward you can create a report file with the name report.html as the example bellow:

snakemake Data_Analysis --report report.html

Visual inspection of the results

For visualization of the results you will need a current version of napari in a new enviroment, please follow the install instructions on their website.

Afterwards you can activate the napari environment

conda activate napari-env

Then install openpyxl library to read/write Excel 2010 xlsx/xlsm/xltx/xltm files.

conda install anaconda::openpyxl

Change your current location to the directory workflow/scripts where 3D_visualization.py is located for script execution.

python 3D_visualization.py

Choose as input :

• Each channel image located at workflow/RESLICE directory
• Results located at workflow/CP_OUT directory