Bioinformatics Exercises
  A.Problems. Protein Structure
  Writer : Seyeon Weon   Updated : 10-14   Hit : 2723   Updates 

(For those who have taken the courses and want to submit for evaluation, please read the instructions linked on the table of contents page.)

  1. Answer the following questions about protein structure:

    (a) In protein structure, alpha helix and beta sheet are the major themes. Why does it have to be that way?
    (b) Explain why the change of dihedral angles plays the major role in protein structure in terms of energy change.
    (c) In protein structure, weak forces are important. Enumerate these weak forces.
    (d) Glycine plays a special role in protein structure. Explain it briefly.
    (e) Between two chiral forms of beta-alpha-beta structure, right handed one is rare. Why is it so?
    (f) Briefly describe your perception about the domains in protein structure.
    (Within 10 lines for each)
  2. Answer the following questions about protein structure prediction:

    (a) Why did PSI-BLAST have a big impact on protein structure prediction?
    (b) What are the major reasons why ab initio prediction is such a hard problem?
    (c) Refining after homology modeling can be done with a reasonable computational power. Why is it so?
    (d) Structural classification and fold recognition are other types of activity in protein structure. What is the major reason that makes these methods work?
    (e) Briefly explain the rationale behind the so-called structural genomics project.
    (f) Structural alignment is of course much harder than sequence alignment. Also, threading is another type of method that makes use of structure. What could we get more with these methods?
    (Within 10 lines for each)
  3. Another method for evaluating the quality of predictions of protein structure is GDT(Global Distance Test), which is {100*(GDT1 + GDT2/2 + GDT4/4 + GDT8/8)}/{4*(number of residues)}. GDT1 is the number of residues superimposed within 1 angstrom, and GDT2 is the number within 2 angstrom, and so on. The residues counted in GDT1 is also counted in GDT2 since they are within 2 angstrom. Therefore, GDT8 also contains all residues counted in GDT1, GDT2, and GDT4. Comparing to RMSD, explain the advantage of GDT.
  4. Below is a distance matrix obtained from the DALI algorithm applied to a protein. Can you tell (roughly) what type of protein it is? Also, describe how the distance matrices may look like for other types of proteins.

  5. Draw the dynamic programming table for the alignment of mass spectrum data of two peptide sequences below:


    Scoring scheme can be any reasonable one and a rough drawing is sufficient. Also, ignore ion types.