Skip to content
Go to file

Latest commit

* Run tests in docker environment for travis

* Remove Makefile

* Use python packages as used previously

* Create a data directory in notebooks folder as well

* Update notebooks after using diabetes.arff

* Change the environment variable

* Put debug true temporarily

Git stats


Failed to load latest commit information.
Latest commit message
Commit time

A data mining suite for gene expression data

Candis is an open source data mining suite (released under the GNU General Public License v3) for gene expression data that consists of a wide collection of tools you require, right from Data Extraction to Model Deployment. candis is built on top of the toolkit - CancerDiscover written by the bioinformaticians at HelikarLab.

Citation: If you use candis please cite our work
Mohammed, A., Biegert, G., Adamec, J., & Helikar, T. (2017). Identification of potential tissue-specific cancer biomarkers and development of cancer versus normal genomic classifiers. Oncotarget, 8(49), 85692-85715.


Mohammed, A., Biegert, G., Adamec, J., & Helikar, T. (2018). CancerDiscover: An integrative pipeline for cancer biomarker and cancer class prediction from high-throughput sequencing data. Oncotarget, 9(2), 2565-2573.

WARNING: candis currently is still in dev mode and not production-ready yet. In case if you run across bugs or errors, raise an issue over here.

Table of Contents


Assuming you've installed dependencies, simply

$ pip install candis


$ curl -sL | python # with dependencies

... and launch candis's development server:

$ candis

To install candis right from scratch, check out our exhaustive guides:

Docker Image

You can also attempt to install candis via Docker as follows:

$ docker pull helikarlab/candis

... and simply run the image optionally mapping the port 5000.

$ docker run -p 8888:5000 helikarlab/candis


After cloning the repository, build from the updated Dockerfile and docker-compose.yml:

For development:

$ ./manage up -d --build

For production:

$ CANDIS_ENVIRONMENT=production ./manage up -d --build

Then go to localhost:5000 in your browser to open the app.

Other Commands:

$ ./manage [service] [command]

$ ./manage db backup			 		# Backup the database
$ ./manage db restore /path/to/backup	# Restore a snapshot
$ ./manage db backups 				 	# List all backups


Launching the RIA (Rich Internet Application)

via CLI

$ candis


$ python -m candis

via Python

>>> import candis
>>> candis.main()

Using the CLI (Command Line Interface)

$ candis --cdata path/to/data.cdata --config path/to/config.json

Using the Jupyter Notebook from inside the docker container

  • Starting the jupyter notebook server inside the candis app container
$ docker-compose exec app jupyter notebook --ip --no-browser --allow-root


  • Converting a CDATA to an ARFF file

     >>> import candis
     >>> cdata ='path/to/data.cdata')

    Then, simply use the CData.toARFF API:

     >>> cdata.toARFF('path/to/data.arff')
  • Running a Pipeline.

     >>> pipe = candis.Pipeline()
     >>> while pipe.status == candis.Pipeline.RUNNING:
     ...     # do something while pipeline is running


  • Production Dependencies
    • R
    • WEKA (NOTE: Requires Java)
    • Python 3.6+ and PIP (Python's Package Manager)
    • NumPy
  • Development Dependencies


Dr. Tomas Helikar

Principal Investigator

Dr. Akram Mohammed

Author and Maintainer

Achilles Rasquinha

Author and Maintainer

Rupav Jain

Author and Maintainer


This software has been released under the GNU General Public License v3.

You can’t perform that action at this time.