Skip to content
A web tool to do automatic spots detections and positional adjustments for Spatial Transcriptomics Datasets.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
client
sample_images Add sample images to the repo Aug 14, 2017
server Bump Sanic version Sep 14, 2018
.gitignore
AUTHORS Update AUTHORS Oct 4, 2017
DEPENDENCIES Added additional libraries to DEPENDENCIES file Oct 12, 2016
LICENSE Added SpinKit and a DEPENDENCIES file for licenses of dependencies Oct 12, 2016
README.md Merge branch 'release-v2.0.0' Jun 15, 2018

README.md

ST Spot Detector

A web tool for automatic spot detection and positional adjustments for ST datasets.

The arrays used to generate ST datasets may contain positional variations due to printing artifacts. This web tool aims to detect correct spot positions using the images generated from the ST protocol. In order to obtain relevant experimental data, it is also possible to automatically select the spots which are located under the tissue, using a corresponding HE image. The spot positions and selections are further adjustable to one's own needs. A file is generated which contains the corrected spot coordinates of the ST data as adjusted array coordinates and pixel coordinates as well as file containing a 3x3 affine matrix to transform array coordinates to pixel coordinates which can be useful for downstream analysis.

Dependencies

The following packages are required:

Python 3.6+
Node.js 10.0+
npm 6.0+

Additionally, the ST tissue recognition library needs to be installed.

A modern browser is required for the front-end interface. The web app has been tested on Chrome 66 and Firefox 60.

Usage

If you want to deploy the ST Spot detector locally on your computer you can use this singularity container. Otherwise follow the deployment instructions below.

Download

  1. Clone the repository

    git clone https://github.com/SpatialTranscriptomicsResearch/st_spot_detector.git
    
  2. Move into the directory

    cd st_spot_detector
    

Server setup

  1. Install the dependencies

  2. Create and activate a Python virtual environment

    pip install virtualenv
    virtualenv venv
    source venv/bin/activate
    
  3. Install the dependencies in requirements.txt.

    cd server
    pip install -r requirements.txt
    
  4. Install the tissue recognition library (still within the Python virtual environment). Follow the instructions here.

  5. Build the client side files:

    cd ../client
    npm install
    make dist
    
  6. Run the server with the following command:

    cd ../server
    python -m app
    

    The server is WSGI-compatible and can, alternatively, be run with a WSGI server of your choice.

  7. Optional. It may be desirable to configure port-forwarding to be able to access the web tool through port 80, e.g. using iptables:

    sudo iptables -t nat -A PREROUTING -p tcp --dport 80 -i eth0 -j DNAT --to 0.0.0.0:8080
    
    

For any queries or concerns, feel free to contact the authors at the addresses given below.

Citation

If you want to cite the ST Spot detector:

ST Spot Detector: a web-based application for automatic spot and tissue detection for Spatial Transcriptomics image data sets DOI: 10.1093/bioinformatics/bty030

Manual

For a guide on using the ST spots detection tool, please refer to this guide.

License

MIT (see LICENSE).

Authors

See AUTHORS.

Contact

Kim Wong kim.wong@scilifelab.se

Ludvig Bergenstråhle ludvig.bergenstrahle@scilifelab.se

Jose Fernandez jose.fernandez.navarro@scilifelab.se

You can’t perform that action at this time.