CellQuant was created for the O'Donnell Lab at the University of Pittsburgh to quantify plasma membrane and vacuolar fluorescence in yeast cells.
- Two membrane detection algorithms for cases with low/high fluorescence at the cell membrane
- Supplemental visual output
- Applet to demonstrate the program at different application settings
- Applet to sort and graph results
-
Install Docker
-
Make sure Docker is running and pull the cellquant image from Docker Hub. You only need to do this once to install the program. Running this command will also install the latest version of the program if a newer version is available.
docker pull odonnelllab/cellquant
-
Run the following command to create a container and run on port 3838. The -v flag maps the container directories with the local file system.
Mac:
docker run --rm -p 3838:3838 -v $HOME:/srv/shiny-server/home odonnelllab/cellquant
Windows:
docker run --rm -p 3838:3838 -v /$HOME:/srv/shiny-server/home odonnelllab/cellquant
CellQuant will be running on localhost:3838.
A detailed installation tutorial is available here.
PMDetectionDemo is a visual tool to help users configure the CellQuant automated pipeline. The applet demonstrates all possible membrane detection algorithms at all possible settings. Installation instructions and source code can be found on the GitHub repository but the Docker commands repeated below:
Install: docker pull odonnelllab/pm-detection-demo
Run: docker run --rm -p 3838:3838 -v $HOME:/srv/shiny-server/home odonnelllab/pm-detection-demo
The applet will be running on locahost:3838.
QuantSort is an applet designed to help sort, group, and graph the output of CellQuant. It can be found online at https://odonnelllab.shinyapps.io/QuantSort/ or downloaded as part of the CellQuant bundled release.
- Docker
CellQuant was built with R 3.6.3 with the following package dependencies:
Bioconductor::EBImage
stringr
shiny
shinyFiles
shinyjs
shinythemes
tidyr
- Mac OS compatibility