Problem Based Learning SoSe 2019

Type 5S (1 year module, 2  SWS/Summer, 3 SWS/Winter)
Lecturer Burkhard Rost, Maria Schelling
Time Monday, 12:00 - 14:00
Room MI 01.09.034
Language English


The topics will be published in the week before the 1st meeting. Two students will work on the same topic, but will focus on different ways to predict the given feature.

Kick-off Meeting

The first meeting will take place in room MI 01.09.034 on Monday, 29 April from 12:30-14:00.


  • 27.5.19: How to give a good presentation? (Presentation by the Rostlab)
  • 17.6.19: Background talks (students presenting: Leonardo/Felix, Isabell/Max, Robin/Faezeh, Jonas/Leo) we will start at 12:00
  • 24.6.19: Background talks (students presenting: Alex/Marc, Kalin/Justin, Leonie/Henrik/Leopold) we will start at 12:00
  • 01.07.19: Introduction to Machine Learning with Python (Presentation by the Rostlab)
  • 15.07.19: Preliminary talks we will start at 12:00
  • 14.10.19: Intermediate talks (students presenting: Leonie, Henrik, Leopold, Leo, Jonas, Robin) we will start at 12:00
  • 21.10.19: Intermediate talks (students presenting: Leonardo, Isabell, Justin, Kalin, Alex) we will start at 12:00
  • 25.11: Final talks (students presenting: Leo, Jonas, Robin)
  • 09.12: Final talks (students presenting: Leonardo, Isabell, Justin, Kalin) we will start at 12:00
  • 16.12: Final talks (students presenting: Henrik, Leopold, Leonie, Alex) we will start at 12:00

Course outline and goals

This course focuses on the application of machine learning to predict various aspects of protein function and structure. During this course, independent of the assigned prediction task, students are going to:

  • perform literature research of a pre-defined topic
  • get a general understanding of machine learning and how to apply machine learning to biological data
  • develop and correctly evaluate a machine learning model including parameter optimization (using Python 3)
  • present milestones and final results in various presentations to the other students and supervisors
  • summarize results in a paper-like scientific report at the end of the course

Students will work in groups of 2 (exception: drug target prediction). They will present their topic and biological background as well as their dataset together. They will work on the same prediction task, but will follow different approaches as discussed with their supervisor. In the end, each student will separately present his/her results in a final talk and in a written scientific report.


  • Prediction of transmembrane helices (supervisor: Michael Bernhofer)
  • Prediction of signal peptides (supervisor: Michael Bernhofer)
  • Protein function prediction using GO terms (supervisor: Michael Heinzinger)
  • Protein function prediction using EC numbers (supervisor: Michael Heinzinger)
  • Localization prediction (supervisor: Maria Schelling)
  • Binding prediction (supervisor: Maria Schelling)
  • Drug target prediction (supervisor: Christian Dallago) this topic will be assigned to 3 students