We're upgrading the GitLab server on Monday August 21 starting at ~ 8:00 in the morning till ~9:30 am , Gitlab will be updated from 9.0.12 to 9.0.13 Please let us know if you notice anything broken afterwards. -- Michel


Programming course

Supporting material for our programming course

Programming course

This repository contains material for the second installment (July 2014) of a programming course for scientists organised by the department of Human Genetics of the Leiden University Medical Center.

The course is targeted at PhD students, Postdocs, or anyone willing to learn how to program in Python. Students are assumed to have some experience with programming, but not necessarily in Python, and the UNIX shell.

The program consists of four mornings with lessons and some assignments to be done in your own time (i.e., during the afternoons).

See the Trac Wiki for more information.


Please note that the above mentioned date and location are subject to change.

Software installation

See the instructions in INSTALL.md.


The top-level directory contains materials for the following lessons:

  1. Welcome (slides)
  2. Introduction to Python (slides 1, slides 2, slides 3)
  3. More Python Goodness (notebook 1, notebook 2)
  4. Working with NumPy arrays (slides)
  5. IPython Notebook (notebook)
  6. Plotting with matplotlib (notebook)
  7. Painting Pandas (slides)
  8. Object-oriented programming (slides)
  9. A sip of Biopython (notebook 1, notebook 2)

As indicated, some of the lessons are slideshows, whereas others are just notebooks we scroll through during class. The links above are all one-page static renderings on IPython Notebook Viewer.

We also have a repository with material for the assignments.


We apply some custom styling to the notebooks (e.g., body width, font), which is loaded in the last cell. This loads styles/notebook.css and styles/notebook.js.

A variant styles/notebook.css.small is provided that is more suitable for use on low-resolution displays. To use it, manually change the reference to this file in the bottom cell, and rerun it.


The sources for the slideshows are also IPython notebooks and you can edit them by starting a notebook server:

ipython notebook

Choose Slideshow in the Cell Toolbar menu.

Some aditional information on editing slides in the Notebook can be found here in this presentation.

We also apply some custom styling to the slideshows, which is loaded in the last cell.

Live rendering of the slides

You can use nbconvert to convert the slides to HTML and serve them. For example:

ipython nbconvert --to slides --post serve numpy.ipynb

This will open the slides in a new browser window. If you don't want that, add this argument:


To serve on another IP address than the default, use the ip configuration of the serve postprocessing. For example, to listen on all IP addresses:


Changing the port can be done similarly with port.

By default, the reveal.js library is loaded over the internet from a CDN. I think it's usually not a good idea to rely on internet connectivity for your slides, so you can also place a copy of reveal.js on your local computer and specify the location like this:

--reveal-prefix reveal.js

This would look for the reveal.js library in the reveal.js directory. A Git submodule is already setup for this, so you can just do:

git submodule init
git submodule update

(Unfortunately, there are other online dependencies such Font Awesome, so without an internet connection, not everything will look ok, but it will work.)

Also, if you just want to compile the slides to HTML without serving them to your browser, leave out the --post serve argument.