At the end of the course, students should have a profound understanding of:
- molecular modelling and its applications in life sciences
- the choices one has to make to model a system properly and how to describe and model interactions between particles
- the modelling techniques used in this field of research, in particular homology modelling, molecular dynamics simulations and biomolecular docking
- modelling of biochemical systems on computers using different software
After completing the course the student is able:
- to choose the best modelling method for a given application
- to generate 3D models of proteins from sequence information
- to study the conformational landscape of molecules with molecular simulations
- to model the interaction between biomolecules
- to use relevant modelling software under a Linux environment
- to critically analyse modelling results
Description of content|
Computational structural biology is a mature field of research whose contribution to life sciences is becoming increasingly more appreciated. The aim of this course is to provide a solid basis of computational structural biology methods, with an emphasis on practical protein modelling and simulation, to interested MSc and PhD students in the life sciences. Further, given the lack of emphasis on practical computational research in MSc and PhD courses, this course is designed to have a smooth learning curve regarding the GNU/Linux environment and its command-line interface. By the end of the course, the students are expected to master the three major computational structural biology methods – homology modelling, molecular dynamics, and protein docking – not only from a user perspective but also from a theoretical standpoint.
The course is scheduled to last two-weeks with theoretical lectures (including some exercises) in the morning (9:00-12:00) and practical sessions in the afternoon (13:15–17:00). The students are required to summarize the results of the computer practicals by writing a short article in the form of a communication for the Journal of the American Chemical Society. The last 2 days of the course are reserved for a guest lecture giving an industry perspective to the topic, the article writing, self study and a final exam. The first afternoon is devoted to the installation of the material and a short crash-course on GNU/Linux and the command-line interface.
The theoretical part consists of classical lectures (see programme above) covering the various aspect of computational modelling of biomolecular systems, together with a few exercises sessions integrated within the lectures. These exercises are meant to illustrate some aspects of the methodology discussed. Through a number of simple python scripts, students will be able to play with some of the techniques discussed, and visualize the impact of various parameters on the simulation results. The material for the lectures is based in parts on the following book (recommended for further in depth reading):
A.E. Leach, Molecular Modelling: Principles and Applications, 2nd edition, Pearson Eduction Ltd, 2001.
PDF of the lecture slides will be provided after each lecture.
The computer practical part is divided in three main modules, each focused on a major computational structural biology method. The philosophy of the practical components of the course follows also our previous experience: the students are given a set of instructions and follow them at their own pace, with the assistants helping out whenever necessary.
The first module comprises the setup and analysis of a molecular dynamics simulation of a small peptide and is based on our previous BSc course and peer-reviewed educational article published in Biochemistry and Molecular Biology Education . The students will make use of GROMACS , a widely used software for molecular dynamics simulation, to characterize the conformational landscape of a small peptide and extract representatives that will be used in the third and last module.
The second module covers homology modelling and guides the students throughout all the stages of the process of building a protein model from a structurally characterized homologue. It makes use of HMMER  for sequence searches, MODELLER  for model building, and Pymol  for visualization. The students will use the programs’ command-line interface instead of the readily available web servers. This, we hope, will familiarize them with an important component of computational research, as well as bring them closer to the tools and their many options.
The third module covers the docking of the homology model built in the second module with the peptide conformers extracted from the simulation of the first module. The students will use bioinformatics interface predictors and HADDOCK web servers  to predict the interface between the two molecules and build models of their interaction by data-driven docking.
- Rodrigues JPGLM, Melquiond ASJ, Bonvin AMJJ (2015). Molecular Dynamics characterization of the conformational landscape of a small peptide. Biochemistry and Molecular Biology Education. In press.