Install and run Cellpose

Software page:

GitHub - MouseLand/cellpose: a generalist algorithm for cellular segmentation with human-in-the-loop capabilities

To install:

Importantly, if you have your own anaconda installation - make sure this is not active.

srun -p interactive --pty --mem 10G -c 1  -t 12:00:0 bash

module purge 

module load miniconda3/23.1.0 

git clone  https://github.com/MouseLand/cellpose.git  

cd cellpose 

mamba env create -f environment.yml 

source activate cellpose

mamba install -y -c anaconda -c conda-forge libxcb xcb-util-wm xcb-util-image xcb-util-keysyms xcb-util-renderutil-cos7-x86_64  qt pyqt python==3.11

pip install PyQt5==5.14.1

pip install .[gui]

Once that finishes, you should be able to request a Desktop Mate session from O2Portal (https://o2portal.rc.hms.harvard.edu). At the top, click on HMS-RC Applications, and navigate to Desktop Mate. Fill out the form, requesting as many resources as you need (or think you need), then start the session. Once it starts, enter the session, then open a terminal at the top left (it's the black box icon), then do the following:

module load miniconda3/23.1.0 
source activate cellpose
cellpose

and after a few seconds, the GUI should pop up.

Please let us know if anything does not work: rchelp@hms.harvard.edu

Note: the above commands are based on Alex’s notes.

HMS IT RC O2

Install and run Cellpose

Analytics

Related content