Structural Bioinformatics

Course content

Overview of protein structure, experimental structure determination and structure analysis, calculation of RMSD and structure superposition, fold classification (CATH, SCOP, new methods), inference of function from structure, 1D prediction (secondary structure, solvent exposure, neural networks, HMMSTR) & sampling of protein conformations (TorusDBN, directional statistics), homology modelling & the loop closure problem, knowledge based energy functions, de novo prediction methods (ROSETTA, LINUS). RNA 2D and 3D structure prediction. Python based practicals using Biopython's Bio.PDB module.

Education

MSc Programme in Bioinformatics
MSc Programme in Biochemistry
MSc Programme in Biology-Biotechnology

Learning outcome

Knowledge:   

  • Know basic aspects of protein physics (folding, physical forces, thermodynamics, statistical physics)
  • Know goals and methods of fold classification and function-from-structure for proteins
  • Know goals and methods of 1D structure prediction and parameterization (secondary structure, backbone structure, solvent exposure)
  • Know basic vector and matrix algebra.
  • Know goals and methods of ab initio and homology modelling of proteins.
  • Know the basics of probabilistic structure prediction
  • Know the RMSD algorithm in mathematical detail
  • To obtain insight in and background for bioinformatic methods for RNA secondary structure prediction
  • Knowledge about structural (2D) RNA alignments.
  • Knowledge of RNA 3D structure and of 3D aspects of RNA structure prediction.
  • Knowledge about the applied aspects of RNA structure prediction methods in genome analysis


Skills:

  • To be able to implement the RMSD algorithm
  • To be able to use the PDB database and PDB files
  • Use databases such as SCOP and CATH
  • To be able to implement structural bioinformatics algorithms in Python and Biopython's Bio.PDB
  • To be able to visualize and analyze biomolecular structures using PyMol
  • To be able to implement the Nussinov algorithm at least to the level of computing the max score for an unbifurcated RNA secondary structure.
  • Apply RNA structure prediction programs


Competences:

  • Analyze a protein structure with relevant methods and algorithms
  • Analyze a problem in structural bioinformatics and outline a matching algorithm
  • Distinguish different types of protein structure prediction and their degree of difficulty
  • How the energy parameters, energy model, probabilistic models provides the core components for RNA structure prediction.
  • Interpret the results of RNA structure prediction methods and relate them to that of prediction on "background" sequence
  • How RNA structure prediction are made more reliable wen using multiple sequences.

Lectures (4-6 pr week), exercises (2-4 hours pr. week)

See Absalon.

Biological sequence analysis or similar. Linux and python programming or similar.

ECTS
7,5 ECTS
Type of assessment
Written assignment, 54 hours
The exam will consist of a take home programming task, that is, the development of a program that solves a specific structural bioinformatics problem. The program needs to be implemented in Phython using Bio.PDB (Biophython´s structural toolkit). You will have time for the take home assignment in week 3 and 4.
Marking scale
7-point grading scale
Censorship form
No external censorship
Several internal examiners/co-examiners.
Criteria for exam assessment

To obtain the grade 12:

  • The student must be able to explain the motivation, biological relevance and use of the methods covered in the course.
  • The student must be able to present and explain the mathematical and algorithmic details of the methods covered in the course.
  • The student must be able to suggest which methods and programs to apply for a given biological problem, and to point out problems and difficulties relating to such applications.
  • The student must be able to implement standard algorithms to solve standard structural bioinformatics problems in Python and Bio.PDB, and to design and implement novel, related algorithms to solve related problems.

Single subject courses (day)

  • Category
  • Hours
  • Exam
  • 54
  • Preparation
  • 91
  • Lectures
  • 28
  • Practical exercises
  • 21
  • Project work
  • 5
  • Guidance
  • 7
  • English
  • 206