Structural Bioinformatics in Micbio 565 - 2008
by Eric Martz
guest lecturer in Steve Sandler's Laboratory in Molecular Genetics
Dept Microbiology, University of Massachusetts (UMass), Amherst

To find this page later: http://565.molviz.org
On Macs, Safari works a little better than Firefox.
On Windows, either Internet Explorer or Firefox is fine.
I. Introduction
II. Review of Protein Structure
III. Examples 		IV. Tools
V. Powerpoint Assignment
    I. Introduction to Structural Bioinformatics

  1. Introduction to Structural Bioinformatics includes:
    1. Why do we care about 3D macromolecular structure?
    2. What are 3D structure data?
    3. Where do 3D structure data come from?
    4. How much 3D structure knowledge do we have?
    5. Primary and Derived 3D Structure Databases

    II. Review of Protein Structure

  2. Central dogma: DNA -> RNA -> Protein
  3. 20 Amino acids
    • Codon = 3 nucleotides; 4 nucleotides3 = 64 codons.
  4. Polypeptide chain geometry and steric restrictions
  5. Covalent and non-covalent chemical bonds
    Typical hydrogen bond within a protein: hydrogen donor atom is covalently bonded to hydrogen; acceptor atom is not.
    • Covalent bonds: lengths and angles nearly constant
    • Non-covalent bonds: variable lengths and angles
      • Salt bridges (up to 4.0 Å in proteins)
      • Hydrogen bonds (2.5-3.5 Å in proteins)
      • Cation-pi orbital interactions (up to 6.0 Å in proteins)
      • van der Waals interactions (up to 4.5 Å in proteins)
  6. Secondary Structure
  7. Folding: hydrophobic collapse
  8. Protein folds cannot be reliably predicted from sequence alone (using ab initio theory).

    III. Examples

  9. "The structure of H5N1 avian influenza neuraminidase suggests new opportunities for drug design", Russell et al., Nature 443:45 (September 7, 2006): on-campus   off-campus

  10. "Crystal structures of oseltamivir-resistant influenza virus neuraminidase mutants", Collins et al., Nature 453:1258 (June 26, 2008): on-campus   off-campus

    IV. Tools

  11. Seeing 3D protein molecules:
    • First, obtain the PDB identification codes (always 4 characters) for the molecules of interest. (See Finding below.) Examples:
      • 1d66: the Gal4 transcriptional regulator binding to DNA.
      • 1bl8: the potassium channel, a trans-membrane protein.
      • 1gzx: oxygenated human hemoglobin.
      • 1sh9: HIV protease bound to the inhibitor Ritonavir. (See also HIV protease.)
    • At Proteopedia.Org, enter the PDB code in the search slot and click Go.
    • For more detailed exploration, while looking at the molecule in Proteopedia, click on the link to FirstGlance under Resources beneath the molecule.

  12. Finding published molecules of interest
    • Browsing:
    • Searching:
      • PDBLite.org   Simple, one-slot search. Clear help.
      • OCA (developed by Jaime Prilusky at Weizmann)   Powerful and straightforward.
      • Research Collaboratory for Structural Bioinformatics (RCSB, USA), part of the World Wide Protein Data Bank (wwpdb.org):
        • One-slot search at main page, pdb.org.
        • Or click Advanced Search at pdb.org   Powerful but difficult to use; help is often inadequate.

  13. Visualizing Protein Evolution:
    Animation made with Polyview-3D. Larger samples and more information...

  14. Publication-Quality Molecular Images
    • Polyview-3D is easy to use.
    • It makes beautiful static images that are ideal for scientific papers or Powerpoint.
    • It also makes animations (rotating molecules) that play in Powerpoint slides or web pages.

  15. Optional 3D Molecular Visualization Resources

    V. Powerpoint Assignment

  16. Select a personal protein molecule from one of the Browsing sites listed above, or by searching for a molecule of interest to you with one of the Searching sites above. Write down its PDB code.
  17. Construct Powerpoint slides with the following contents. Save your slides with the filename your-last-name_565.ppt, e.g. sandler_565.ppt. When completed, email your completed .ppt file to
    (Your slides need to be clear and well organized, but not polished or beautiful. They do not need to be presentation quality, but rather they are to serve as a notebook recording your work.)
  18. Sample completed assignment.

    This is not a test. It is to help you learn by doing. Ask for help!

      Slide 1:
    • Your name. Your molecule's name and four-character PDB identification code. The experimental method (X-ray diffraction or Nuclear Magnetic Resonance?) and resolution (if X-ray) or number of models (if NMR).
      (Proteopedia gives the resolution beneath the molecule for X-ray crystallographic results, or the number of models for NMR results.)
    • How many polymer chains (protein or nucleic acid chains) are in the published PDB file (the asymmetric unit)? (Both Proteopedia and FirstGlance obtain the published PDB file from the RCSB Protein Data Bank.)
      (If the number of chains is not obvious, in FirstGlance in Jmol, click on the PDB button, or in Proteopedia, click on RCSB under Resources. At RCSB-PDB, click on the Biology and Chemistry Tab.)
    • A snapshot of your molecule from FirstGlance in Jmol. How to take a snapshot is also linked at the bottom left (scroll down) in FirstGlance in Jmol.
      Be sure to clearly label all snapshots and animations in your slides.

      Slide 2:
    • How many polymer chains are in the "biological molecule" according to the Probable Quaternary Structure server?
      (In FirstGlance in Jmol, click on the PQS button.)
    • Side-by-side snapshots (in FirstGlance in Jmol) of the published PDB file from RCSB, and the result at PQS.
      (Even if they have the same number of chains, they may not be in the same conformation. More..)
      To display a PQS result in FirstGlance so you can take a snapshot:
      1. Macs: Control-Click (Windows: Right-Click) on the RasMol link to a filename ending ".mmol", then Copy Link/Link-Location/Shortcut.
      2. Go to firstglance.jmol.org.
      3. Click on "enter a molecule's URL".
      4. Paste the URL into the slot.
      5. Click "Submit".

      Slide 3:
    • 3-letter abbreviations and full names for all ligands. If none, so state.
      (Proteopedia lists the 1 to 3-letter abbreviations for each ligand under the molecule. Click on each one to see its full name shown in red at the bottom of the molecule.)

      Slide 4:
    • From FirstGlance, a snapshot showing the Hydrophobic/Polar view of your protein.
      Is there a large hydrophobic patch? If so, show it in your snapshot. Do you think your protein is soluble? Compare it to an insoluble (transmembrane) protein, such as porin. If your protein is a membrane protein, show the snapshot from Orientations of Proteins in Membranes.

      Slide 5:
    • From the Charge.. view in FirstGlance, a snapshot showing the positive and negative charges on your protein.

      Slide 6:
    • A snapshot of the ConSurf view (in FirstGlance) of one protein chain in your molecule.
      If there is a highly conserved or variable region, show it in your snapshot, and suggest what its function might be.

      How to submit your molecule to ConSurf:
      1. Go to consurf.tau.ac.il.
      2. Enter your PDB ID code.
      3. Enter the one-letter chain identifier to specify which chain to process.
      4. Enter your email address.
      5. Click SUBMIT.
      If you get lots of yellow amino acids, ask for help.

      Slide 7:
    • An animated (rotating) view of your molecule.
      Minimal steps to make an animation:
      1. Use the Firefox browser. (Initial orientation, Set by Jmol does not work in Safari.)
      2. Go to Polyview-3D.
      3. Enter your PDB ID.
      4. Change "Type of request" from "Single slide" to "Animation".
      5. Click "Get Preview".
      6. Under "Extended Settings" (near the bottom) check "Animation Settings".
      7. In the "Animation Settings" box that opens, set Delay to 10/100.
      8. Change "Angle step" to 5 degrees.
      9. Check "Rocking".
      10. Change "angle range" for rocking to 45 degrees.
      11. Click "Get 3D Image".

    • The above steps are the minimum for an animation that avoids putting a heavy load on the server. Feel free to try other options, but while the class is in session, please don't make a large (>300 pixel) animation, or increase the angle range, or decrease the angle step size. Otherwise, the server may get overloaded and take a very long time to produce results. After class is over, feel free to submit more demanding jobs. If you highlight specific residues, please explain why.

      In Powerpoint, animations move only when the slides are projected (full-screen).

      With Windows Powerpoint, you can simply drag the animation directly from the Polyview-3D web page and drop it into a Powerpoint slide.

      With Mac Powerpoint, only the method below produces a slide that will animate continuously -- but animation may work only when the .ppt file created with Mac Powerpoint is projected in Windows Powerpoint.
      1. Control-Click on the animation in the Polyview-3D web page, and select Save Image As ...
      2. Save the image to the Desktop.
      3. Drag the image file (filename ending in .gif) from the Desktop and drop it into a Powerpoint slide.
      4. As stated above, the animation will work only when the saved .ppt file is projected in Windows Powerpoint.


      Slide 8:
    • Optional: A snapshot showing some of the noncovalent bonds to one ligand or other moiety.
      Use the Contacts.. dialog of FirstGlance in Jmol. If there are no ligands, use contacts between chains, or if only one chain, contacts to a single helix or beta-strand, or to one or a few residues of special importance, such as in a catalytic site. Be sure to state what you selected as a target for the contacts in this snapshot, and make a brief comment about the contacts you show.

Thanks to Frieda Reichsman (moleculesinmotion.com) for creating some of the excellent interactive animations above.