The GenePattern Notebook environment gives GenePattern users the ability to interleave text, graphics, and code with their GenePattern analyses to create "notebooks" that can be edited, shared, and published. GenePattern Notebooks are built on the Jupyter Notebook system (formerly called IPython Notebook) and extend it so that users can take advantage of its ease of use and ability to encapsulate an entire scientific research narrative, without the need to write code.
Full installation instructions for casual use are detailed on the GenePattern website.
The installation instructions below are intended for developers who want to install the project from PIP or GitHub for purposes of software development.
In order to get the GenePattern Notebook working you will first need to install a compatible version of Python. This means you will need either Python 2.7 or Python 3.4+.
Note for Mac Users: Mac comes with Python, but - depending on the version OSX you have - you may have to update to a new version. To do this I would recommend using a package manager such as Homebrew or MacPorts.
You may wish to install the Anaconda distribution of Python from Continuum Analytics. This is a scientific version of Python that ships with many of the most popular Python packages for science, math and data analysis (ex: NumPy, SciPy, Pandas, Matplotlib, IPython, etc.).
Install PIP (Python Package Index) if not already installed (PIP may come with Anaconda distribution; see https://pip.pypa.io/en/latest/installing.html). This is Python's preferred package management system.
Now you should choose to either install GenePattern Notebook from GitHub or from PIP. If you don't know which method you would prefer, then we recommend doing the PIP installation. Then skip to the appropriate step below.
Copy the contents of genepattern-notebook/extension to your development computer and ensure that the resulting directory if on your Python path. To test this, open Python and try to import genepattern. If this is successful, you have a copy of the extension available.
If you don't already have Jupyter installed, you can install it from PIP by running:
pip install jupyter
From here go to the "Load the GenePattern extension" step below.
The easiest way to install GenePattern Notebook is through PIP. It can be installed by executing the following command:
pip install genepattern-notebook
Before you the GenePattern extension for the very first time, you should make sure that it is enable in Jupyter. To do this run the following on the command line:
jupyter nbextension enable --py widgetsnbextension jupyter nbextension install --py genepattern jupyter nbextension enable --py genepattern jupyter serverextension enable --py genepattern
Finally, you may launch Jupyter Notebook by issuing the following command at the terminal:
jupyter notebook
This will start up the notebook kernel and launch your web browser pointing to the Notebook.
If you want to update GenePattern Notebook to a more recent version on PIP, run the following command:
pip install -upgrade --no-deps genepattern-notebook
The current version of the code only works with GenePattern 3.9.3 and up!
Users using the GenePattern Notebook with an older version of GenePattern (3.9.3 or 3.9.4) may need to log into the GenePattern UI before making use of the notebook. The server status message and child jobs will also be unavailable. If you are using one of these older versions, we recommend that you upgrade to the latest version of GenePattern.
Most common GenePattern features are supported in the GenePattern Notebook environment. A few, however, have yet to be implemented. GenePattern features that are not yet supported include:
- Batch job submission
- GenomeSpace integration
- Dynamically updated choice parameters
- Parameter groups