History of X-Ray Diffraction and 3D Structure Determination

History of X-Ray Diffraction

• 1895: X-rays discovered by Roentgen
• 1914: First diffraction pattern of a crystal made by Knipping and von Laue
• 1915: Theory to determine crystal structure from diffraction pattern developed by Bragg
• 1953: DNA structure solved by Watson and Crick
• Present day: Diffraction improved by computer technology; methods used to determine atomic structures and in medical applications

What is X-Ray Diffraction

• Developed by English physicists Sir W.H.Bragg and his son Sir W.L.Bragg in 1913
• Explains why cleavage faces of crystals appear to reflect X-ray beams at certain angles of incidence (theta, q)
• Related to X-ray wave interference
• Direct evidence for the periodic atomic structure of crystals

Crystal Structure

• Crystalline materials characterized by orderly periodic arrangements of atoms
• Unit cell is the basic repeating unit that defines a crystal
• Crystallographic planes are identified by Miller indices

Template Identification

• Search with sequence: Blast, Psi-Blast, and fold recognition methods
• Use biological information: functional annotation in databases, active site/motifs

Alignment

• Initial alignment
• Improve alignment
• Backbone generation
• Loop modeling
• Side chains
• Refinement
• Validation

Template Quality

• Selecting the best template is crucial
• Most successful method in CASP6 was SCWRL by Dunbrack et al.
• Graph-theory knowledge-based method to solve the combinatorial problem of side chain modeling

Side Chains - Accuracy

• Prediction accuracy is high for buried residues, but much lower for surface residues
• Experimental reasons: side chains at the surface are more flexible
• Theoretical reasons: much easier to handle hydrophobic packing in the core than the electrostatic interactions, including H-bonds to waters

Refinement

• Energy minimization
• Molecular dynamics
• Big errors like atom clashes can be removed, but force fields are not perfect and small errors will also be introduced

Error Recovery

• If errors are introduced in the model, they normally cannot be recovered at a later step
• The alignment cannot make up for a bad choice of template
• Loop modeling cannot make up for a poor alignment

Validation

• Most programs will get the bond lengths and angles right
• Ramachandran plot of the model usually looks pretty much like the Ramachandran plot of the template
• Inside/outside distributions of polar and apolar residues can be useful
• Biological/biochemical data: active site residues, modification sites, interaction sites

ProQ Server

• Neural network-based predictor that predicts the quality of a protein model
• Optimized to find correct models in contrast to other methods that are optimized to find native structures