Introduction to Structural Bioinformatics

1
Introduction to Structural Bioinformatics

• Bruce Byrne, PhD
• Sandhya Kortagere, PhD
• Fundamentals of Bioinformatics
• Fall, 2007

2
Administrative Details

• Grades and Wiki.
• Review Phylogenetics assignment from last week.
• Next week is on line no in-class sessions
• We will briefly go over some of the things you
  will be doing for that section at the end of
  class today.
• That exercise will also cover materials we are
  talking about today. No specific assignment is
  due for todays class.
• Review subsequent weeks subjects, assignments
  and mode of delivery (web/classroom)

3
Structural Bioinformatics

• Why might it be useful to understand both the
  sequence and structure of a protein?
• Better understanding of functional interactions
• Protein/ligand binding
• Catalysis
• Insight into aberrant biological phenomena
  (disease)
• Cancer, diabetes, etc
• Develop drugs
• Control disease by interfering with key pathways
  at molecular level
• Influence disease course with metabolic
  inhibitors/activators
• Manipulate gene encoding the protein toward
  altered function
• Gene therapy
• Biopharmaceuticals

4
Pattern Recognition

• Bioinformatics looks for patterns
• Sequences
• Structures
• Identities and Similarities
• When are these characteristics of two objects
  needed and useful in everyday life?

5
Biological Pattern Recognition

• Texts and Sequences
• What patterns are sought in text searches, like
  PubMed?
• How is that varied in a BLAST search of
  biological sequences?
• Expression Data
• How does the color and intensity of signals from
  one experimental condition compare to another in
  the DNA microarray.
• Wait for Functional Genomics unit.
• Molecular Structures
• What common attributes are shared by a ligand?
• Ligand-based design
• How can a characteristic set of amino acids,
  arranged in 3-D, account for ligand or receptor
  binding?

6
What we will be doing

• Today well consider
• How we come by structural information for
  proteins
• The nature of a structure protein file
• Introduce a tool that lets us visualize a
  structure file
• Investigate next week, independently
• Use the NCBI suite of structural tools and
  databases
• Gain skills with a molecular viewer, Cn3D
• How to find and visualize
• Single proteins
• Similar proteins

7
What We Wont Be Doing (yet)

• Manipulating sequence or altering structure
• Modeling a similar sequences with unknown
  structure to a known structure
• Docking a macromolecule and ligand
• Predicting structure based solely on sequence
• ab initio methods

8
Ab initio methods

• Deals with building a peptide structure with just
  sequence information.
• Can have more than one low energy model.
• Iterative process and convergence is a must
• Feasible for short peptide sequences (best model
  was 112aa by Robetta server (David Baker, Nature,
  2007)
• Useful to build loop regions, missing protein
  segments in low resolution structures.
• Complimentary to NMR and X-ray crystallography.

9
General work flow
Source Jayaram B Nucl acid res
2006,34(21),6195-6204
10
Some de novo/Ab Initio programs

• ROBETTA (http//robetta.bakerlab.org) De novo
  Automated structure prediction analysis tool used
  to infer protein structural information from
  protein sequence data.
• PROTINFO (http//protinfo.compbio.washington.edu)
  De novo protein structure prediction web server
  utilizing simulated annealing for generation and
  different scoring functions for selection of
  final five conformers.
• SCRATCH (http//www.igb.uci.edu/servers/psss.html)
  Protein structure and structural features
  prediction server which utilizes recursive neural
  networks, evolutionary information, fragment
  libraries and energy.
• ASTRO-FOLD Astro-fold first principles tertiary
  structure prediction based on overall
  deterministic framework coupled with mixed
  integer optimization.
• ROKKY (http//www.proteinsilico.org/rokky/rokky-p/
  ) De novo structure prediction by the simfold
  energy function with the multi-canonical ensemble
  fragment assembly.
• BHAGEERATH (http//www.scfbio-iitd.res.in/bhageera
  th) Energy based methodology for narrowing down
  the search space of small globular proteins.

11
Methods to Determine Structure

• Solution NMR Nuclear Magnetic Resonance
• Does not require crystallization
• Works on molecules in aqueous solution
• Lower resolution
• X-Ray Crystallography
• 80 of PDB entries
• High resolution
• Large quantities of crystallized protein
• Validity for some biochemical microenvironments?
• Highly hydrated crystals
• Good empirical agreement between solution NMR and
  crystallography

12
Imaging - General

• Ability to create an image related to wavelength
  of energy source
• Light
• X-rays
• Neutron beams
• Electrons
• Lenses focus some energy sources
• Glass light
• Magnets electrons
• Diffraction
• Regular, interpretable patterns resulting from
  the interference of waves

Images Wikipedia
13
X-Ray Crystallography

• Not just an imaging technique
• Data gathering
• Substantial interpretation
• How does it work?
• Wavelength (Ångström range, 10-8 cm ) will cause
  scatter by electron cloud of similar sized atom
• Generally yields a unique model
• Cannot resolve the positions of hydrogen atoms
  unless by modeling or resolution beyond about 1.2
  Å
• Terminal side-chain atoms uncertain for Asp, Gln
  and Thr requires inferred identity

14
Solution NMR

• Analyzes proteins in solution
• Especially useful for smaller proteins, lt 30 kD
• Very important because
• some proteins resist crystallization
• Yields the positions of some hydrogen atoms
• Solution NMR often yields multiple models, in
  comparison with crystallography
• Especially useful in the analysis of large
  complexes

15
The Nature of the Crystal

• Crystal must be
• Single
• A few tenths of a mm in each direction
• Jelly-like protein crystals are
• Fragile and sensitive
• Bound by weak hydrogen bonds, salt bridges and
  hydrophobic interactions
• Contain 50 solvent in channels between stacked
  molecules
• Jelly-like nature permits soaking crystals in
  metal solutions or enzyme inhibitors

16
Obtaining a Crystal

• High concentration, purified protein (2-50 mg/ml
  )
• Add agents to reduce solubility, without
  precipitation
• Evaporate agent from reservoir into hanging drop
  with protein
• Experiment trial and error

17
The Experimental Set-up

• Rotating anode X-ray generator
• Monochromator or focusing mirrors yield single
  wavelength
• Crystal can be repositioned using goinometer
• Photo-plate or electronic recording of diffracted
  pattern

18
Interpretation Theory Bragg's Law

• n? 2d sinT (1)
• Derived by Sir W.H. Bragg and his son Sir W.L.
  Bragg in 1913
• Explains why crystals reflect X-ray beams at
  certain angles of incidence (theta, q).
• Direct evidence for the periodic atomic structure
  of crystals postulated for several centuries.
• The Braggs were awarded the Nobel Prize in
  physics in 1915.

19
Interpretation Practice

• Obtain diffraction pattern
• Position and intensity apparent in image
• Phases of the waves which formed each spot must
  also be determined
• irradiate two or more derivatives of the same
  crystal which differ only in the presence of
  heavy metal ions
• Use multiple wavelengths
• Position, density and phase constitute a
  structure factor.
• PDB structure factor data files permit creation
  of a complete electron density map
• 25 of current PDB files

20
Zinc Fingers (1)

• Well-understood structure with important
  biological function
• Independently folded domain of many proteins
• Requires 1 or more Zinc ions
• A series of Zinc Fingers recognizes specific DNA
  sequences
• Matches regulatory proteins like transcription
  factors

21
Zinc Fingers (2)

• Very common DNA-binding motif
• Characterized by two anti-parallel beta strands
  followed by an alpha helix
• Stabilized by Zn ion interacting with conserved
  histidine (H) and cysteine (C) residues.

22
Search Strategies for Structures

• Use Entrez to look for structure
• Search term zinc finger

• Use TaxBrowser to focus on Mouse
• Select for structures
• Search for zinc finger

23
Examine the Tabs

• Tabs sort your result depending on the
• Search strategy
• Characteristics of the database
• Structure searches
• NMR
• X-ray

24
MMDB Summary Page (1)

• Note layout and features
• Reference
• MMDB and PDB

25
MMDB Summary Page (2)

• Chains (Proteins and Nucleotides, in this
  example)
• 3d Domains
• Domain Families
• you will use more of these as you complete next
  weeks exercise

26
Finding The PDB File

• PDB http//www.rcsb.org/pdb/home/home.do
• Compare and contrast the two organizations
  presentation of the same structure

27
Looking at the PDB Data File

• Note Display File options
• Select PDB File
• Note file structure and similarities to GenBank
• HEADER
• TITLE
• COMPND
• SOURCE
• KEYWDS
• EXPDTA etc
• REMARK
• DBREF
• SEQRES
• etc
• How do these data compare to what you have
  learned about X-ray crystallography?

28
Cn3D The NCBI Viewer

• Demonstration
• Getting the software
• Look at the functions

29
Next Week Beyond

• On-line
• Find structures at NCBI using Entrez tools
• Use Cn3D Viewer to visualize structures
• Use similarity searching tools to find similar
  structures
• Review remaining schedule of classes.