TissUUmaps 2.3.0

TissUUmaps is a lightweight viewer that uses basic web tools to visualize gene expression data or any kind of point data on top of whole slide images.

FlaskTissUUmaps

FlaskTissUUmaps is a minimal python server for TissUUmaps using Flask.

Differences with TissUUmaps

Image format

FlaskTissUUmaps allows to visualize all images from a folder and sub-folders in TissUUmaps. By using a minimal deepzoom server, FlaskTissUUmaps removes the need for creating DZI files of every image.

FlaskTissUUmaps can read whole slide images in any format recognized by the OpenSlide library:

• Aperio (.svs, .tif)
• Hamamatsu (.ndpi, .vms, .vmu)
• Leica (.scn)
• MIRAX (.mrxs)
• Philips (.tiff)
• Sakura (.svslide)
• Trestle (.tif)
• Ventana (.bif, .tif)
• Generic tiled TIFF (.tif)

plus classical images in any format recognized by the PIL library (png, jpg, etc.)

:warning: Warning: classical non pyramidal images will have to be loaded in RAM and will be read entirely for each generated tile. If you have big images, consider converting them in pyramidal format using VIPS.

Saving TissUUmaps state

FlaskTissUUmaps allows to save all TissUUmaps states (Gene expressions, Cell morphology, Regions, Layers) so that you can reload images with all additional information, just as you saved them.

Installation

Note that steps 1-4 are optional and can be replaced by installing a recent version of Python.

1. Install Anaconda (or miniconda).

2. Create a new conda environment from an Anaconda prompt:

   conda create --name tissuumaps
   

3. Activate the new conda environment:

   conda activate tissuumaps
   

4. You should be in the tissuumaps environment:

   (tissuumaps) C:\
   

5. Install the Openslide library

   • For windows:

     • Install Microsoft Visual Studio Build Tools.
     • Install the openslide library.
     • Make sure the bin directory of openslide is in the PATH environment variable.
     • If the libopenslide-0.dll still fails to load, see fix here.

   • For linux:

     • Install openslide using your distribution package (for example in Ubuntu : apt-get install openslide-tools).

   • For MacOs:

     • Install openslide using your MacPorts or Homebrew (port install openslide or brew install openslide).

6. Install python dependencies (openslide, flask and pillow):

   pip install openslide-python flask pillow
   

   In linux, check that you only have one installation of pillow:

   sudo pip uninstall pillow
   pip install pillow
   

7. Clone the FlaskTissUUmaps git repository or download in zip format and extract to a FlaskTissUUmaps folder

   git clone https://github.com/wahlby-lab/FlaskTissUUmaps
   

8. Go to the FlaskTissUUmaps folder and start the FlaskTissUUmaps server:

   cd \Users\myUser\Documents\FlaskTissUUmaps
   python flasktissuumaps.py path_to_image_folder
   

9. Open http://127.0.0.1:5000/ in your favorite browser.

   :warning: Remember that Flask is running on a built-in development server (flask run) and should not be used in production. If you want to deploy FlaskTissUUmaps on a production server, please read https://flask.palletsprojects.com/en/1.1.x/tutorial/deploy/ or any similar tutorial.

Options

FlaskTissUUmaps can be used with the following options:

Usage: flasktissuumaps.py [options] [slide-directory]

Options:
  -h, --help            show this help message and exit
  -B, --ignore-bounds   display entire scan area
  -c FILE, --config=FILE
                        config file
  -d, --debug           run in debugging mode (insecure)
  -e PIXELS, --overlap=PIXELS
                        overlap of adjacent tiles [1]
  -f {jpeg|png}, --format={jpeg|png}
                        image format for tiles [jpeg]
  -l ADDRESS, --listen=ADDRESS
                        address to listen on [127.0.0.1]
  -p PORT, --port=PORT  port to listen on [5000]
  -Q QUALITY, --quality=QUALITY
                        JPEG compression quality [75]
  -s PIXELS, --size=PIXELS
                        tile size [254]
  -D LEVELS, --depth=LEVELS
                        folder depth search for opening files [4]