Quick start

  1. Ensure Nextflow and either Singularity or Docker are installed on your system.

  2. Create a suitable working directory for the PHLEX analysis.

    $ mkdir PHLEX_testing
$ cd PHLEX_testing

  3. Clone the TRACERx-PHLEX repository from github:

    $ git clone --recursive git@github.com:FrancisCrickInstitute/TRACERx-PHLEX.git

  4. Run the TRACERx-PHLEX setup pipeline to download the required weights, test images and shell script to launch TRACERx-PHLEX:

    $ nextflow run TRACERx-PHLEX/PHLEX_setup.nf -w scratch

    Tip

    The directory structure should now look like this:

    PHLEX_testing/
├── TRACERx-PHLEX
│   ├── deep-imcyto
│   ├── docs
│   ├── LICENSE
│   ├── README.rst
│   ├── Spatial-PHLEX
│   └── TYPEx
├── deep-imcyto_weights
│   ├── AE_weights.hdf5
│   ├── boundaries.hdf5
│   ├── com.hdf5
│   ├── nuclear_morph_scaler.pkl
│   └── nucleus_edge_weighted.hdf5
├── PHLEX_test_images
│   ├── P1_TMA006_L_20190619-roi_24.ome.tiff
│   └── P1_TMA007_L_20190619-roi_13.ome.tiff
└── runPHLEX.sh

  5. Launch TRACERx-PHLEX by calling the runPHLEX.sh script from the command line:

    $ ./runPHLEX.sh

  6. TRACERx-PHLEX will then begin.

Note

The first time PHLEX is run, Nextflow will download all the necessary docker images that are required to run PHLEX. This may take a while and it may appear as if a particular process is hanging wilst the containers are built. Subsequent runs will be much faster.

Tutorial

Detailed description on parameters, input files and functionalities is described for each module

PHLEX: deep-imcyto

A module devoted to performing accurate nuclear and cellular segmentation and single cell measurement in multiplex images.

PHLEX: TYPEx

A module for cellular phenotyping from marker expression intensities derived from multiplex images.

Spatial-PHLEX

A module for performing several types of automated spatial analysis.

Contact

mihaela.angelova@crick.ac.uk alastair.magness@crick.ac.uk emma.coliver@crick.ac.uk katey.enfield@crick.ac.uk