Structural Bioinformatics

1
Structural Bioinformatics

• Motivation
• Concepts
• Structure Solving
• Structure Comparison
• Structure Prediction
• Modeling Structural Interaction

2
Motivation

• Holy Grail Mapping between sequence and
  structure. Structure F(Sequence)
• Structure allows the detailed understanding of
  function
• Science-Fiction Simulate life as a movie of
  molecular interactions

3
Concepts

• Protein and DNA structure of maximum interest,
  though growing interest in carbohydrates
• Central Dogma
• Primary structure gt Secondary structure,
  Tertiary structure gt Function
• Primary structure
• The actual permutation, i.e., the sequence per se
• 20L possibilities, but only about 103-4 families
  (equivalence classes)
• High spatial locality
• Secondary structure
• Helix, Sheet, Loop, Coil
• Intermediate spatial locality Local organization
• Tertiary structure
• Only so many folds
• Evolutionary bias?
• Only so many stable structures?
• Low spatial locality The actual 3D structure

4
Structural Bioinformatics

• Macromolecular complex gt
• Protein gt (No. of species 104)
• Fold gt (103)
• Domain gt (103-4)
• Module gt
• Motif gt
• Residue gt
• Atom

Level of abstraction
5
Primary structure

• Determined
• Experimentally
• Sequencing
• Computationally
• Proteome prediction from genome
• Finite number of real-world families based on
  sequence similarity
• Significance Sine qua none
• Databases Swiss-Prot, PIR, Genpept

6
Secondary structure

• Determined
• Experimentally
• Circular dichroism, NMR, Raman spectroscopy
• Computationally (Next semester!)
• Sliding window context analysis
• Periodicity analysis
• Significance
• Higher order building block
• Mechanistic significance in protein folding

7
Tertiary structure

• Determined
• Experimentally
• X-ray crystallography, NMR
• Computationally
• Based on similarity to known structures (Homology
  modeling)
• a priori
• Significance
• Level of abstraction highly indicative of
  function
• Databases PDB, SCOP, CATH, FSSP

8
Structural representation

• Cartesian coordinates (x,y,z) for each atom in
  structure One vector for each atom
• Internal coordinate representation (edges,
  angles) Set of inter-atom vectors
• Visualized in graphical programs as
• Surface representation
• All atoms or just backbone (N-C-C) atoms
• Cartoons Helix, Sheet, Loop shorthand

9
Structure Solving X-ray crystallography(With
apologies to crystallographers)

• Given primary structure, find the tertiary
  structure experimentally
• X-ray crystallography
• Based on property of electrons to scatter X-rays
• Make big beautiful protein crystal (Like sugar
  crystals Slow desiccation of protein soup)
• Shoot X-rays through crystal and record
  diffraction pattern (Ripples from different
  sources forming interference pattern Light waves
  too gross for measurement of small interatomic
  distances)
• Provided phase known, structure can be
  reconstructed using Fourier transforms

10
Structure Solving X-ray crystallography(With
apologies to crystallographers)

• Phase information deduced by
• Incorporation of heavy metals into proteins
  (Crystals from immersion of protein in heavy
  metal soup)
• Use of known structures of similar sequences
  homology based structure solving
• Iterative structure refinement based on matching
  simulated data from model and experimental
  diffraction pattern obtained
• Limitation
• No crystals, no structure!
• Structure is an average over prolonged period of
  time

11
Structure Solving Nuclear Magnetic Resonance
Spectroscopy(With apologies to NMR
spectroscopists)

• Basis
• Based on magnetic properties of spinning protons
  and neutrons
• Different atoms in different (structural)
  contexts react differently to magnetic fields
• Key elements
• Specially prepared (different forms of N and C)
  protein (soup) solution subjected to strong
  electromagnetic field to obtain spectroscopic
  data
• Spectroscopic data interpreted as constraints on
  distances between atoms
• Simulated annealing used to find best match
  between constraints and model

12
Structure Solving Nuclear Magnetic Resonance
Spectroscopy(With apologies to NMR
spectroscopists)

• Limitations
• Set of structures (not unique single solution)
  determined
• On average, higher error than crystallography
• Solves smaller structures than crystallography
• Bright side of structure-solving
• Algorithms and software well developed
• More powerful sources of X-rays (can work with
  smaller crystals and in smaller time periods) and
  electromagnets becoming available
• Optimized centers exist that can do a structure a
  week!
• Challenges Higher resolution (The elusive but
  all-powerful hydrogen bond), Structures of
  complexes (Life is molecular interaction.
  Period)