Install CellProfiler on Centos 7.6 using conda.
March 7, 2019

Install CellProfiler on Centos 7.6 using conda.

Posted on March 7, 2019  •  3 minutes  • 447 words  •  Suggest Changes

CellProfiler is not an easy to install tool; or perhaps I was clumsy on the first attempt (building from source) but I could not get it to work properly on a Linux machine; After another attempt using miniconda, I managed to get it running. This is just a documentation of how I got it working. In case I have to do it again. By no means am I an expert on the case.


I don’t know if these are all the dependencies but at some point I had to install them.

yum install java-1.8.0-openjdk java-1.8.0-openjdk-devel bzip2 mariadb-devel libstdc++-devel gcc-c++ gtk2 ImageMagick ImageMagick-devel

Another issue is that the installation requires is which can not be found, however jbigkit-libs provides; which can be soft-linked and it will then work;

cd /usr/lib64/
ln -s

Install Miniconda

Get the latest version for Linux 64bit (or other) and make it executable; On Centos python 2.7 is default, and I believe it is required for CellProfiler.

chmod 755

Install CellProfiler

Finally we are ready to try the installation; Create a directory “cellprofiler” (not specific)

mkdir cellprofiler
cd cellprofiler
touch environment.yml

environment.yml should contain (using nano, vi, vim, emacs,…)

# run: conda env create -f environment.yml
# run: conda env update -f environment.yml
# run: conda env remove -n cellprofiler
name: cellprofiler
# in order of priority: highest (top) to lowest (bottom)
  - anaconda
  - goodman # mysql-python for mac
  - bioconda
  - cyclus # java-jdk for windows
  - conda-forge # libxml2 for windows
  - BjornFJohansson # wxpython for linux
  - appdirs
  - boto3
  - cython
  - h5py
  - ipywidgets
  - java-jdk
  - joblib
  - jupyter
  - libtiff
  - libxml2
  - libxslt
  - lxml
  - packaging
  - pillow
  - pip
  - python=2
  - pyzmq=15.3.0
  - mahotas
  - matplotlib!=2.1.0,>2.0.0
  - mysqlclient
  - numpy
  - raven
  - requests
  - scikit-image>=0.13
  - scikit-learn
  - scipy
  - sphinx
  - tifffile
  - wxpython=
  - pip:
    - cellh5
    - centrosome
    - inflect
    - prokaryote==2.4.0
    - javabridge==1.0.15
    - python-bioformats==1.4.0
    - git+


After this, make the environment; using :  (will take a while)

conda env create -f environment.yml

While debugging you can also use : (to update)

conda env update -f environment.yml

This kinda works but generates two warnings that don’t seem to impact the tools (but perhaps I haven’t use specific functions that depend on these)

cellprofiler 3.1.8 has requirement prokaryote==2.4.1, but you'll have prokaryote 2.4.0 which is incompatible.
cellprofiler 3.1.8 has requirement python-bioformats==1.5.2, but you'll have python-bioformats 1.4.0 which is incompatible.

Once that is finished, we can activate & run.

conda activate cellprofiler

Since this is on a server, I also needed to allow X11 forwarding over ssh;


If you enjoyed this website, consider buying me a Dr. Pepper

Buy me a Dr PepperBuy me a Dr Pepper