The 8th Protein Folding School - PowerPoint PPT Presentation

1 / 20
About This Presentation
Title:

The 8th Protein Folding School

Description:

The 8th Protein Folding School. 1?. ??? (KIAS) ??? (Seoul National Univ.) ??? ... 1. The ab initio loop prediction. 2. Database approach to loop prediction ... – PowerPoint PPT presentation

Number of Views:25
Avg rating:3.0/5.0
Slides: 21
Provided by: YunJ7
Category:
Tags: 8th | ab | dihedral | folding | in | initio | protein | school

less

Transcript and Presenter's Notes

Title: The 8th Protein Folding School


1
The 8th Protein Folding School
1? ??? (KIAS) ??? (Seoul National Univ.) ???
(Yonsei Univ.)
2
Step in comparative protein Structure modeling
3
(No Transcript)
4
(No Transcript)
5
Searching for suitable template structures
  • Three classes of protein comparison
    method
  • Pairwise sequence-sequence comparisons. (BLAST)
  • Multiple sequence comparisons. (PSI-BLAST)
  • Threading or 3D template matching method.

6
Template selection
  • Selection appropriate templates
  • The family of proteins.
  • The template environment (solvent, pH, ligands).
  • The quality of the experimental template
    structure.

7
Target-Template Alignment
  • Identity over 40 almost always correct.
  • Less sequence similarity CLUSTAL.
  • More difficult alignment cases-twilight zone
  • multiple structure and sequence information.

8
Model Building
1. Modeling by assembly of rigid bodies. 2.
Modeling by segment matching or coordinate
reconstruction. 3. Modeling by satisfaction of
spatial restraints.
These homology derived restraints are usually
supplemented by stereochemical restraints on bond
lengths, bond angles, dihedral angles, and
nonbonded atomatom contacts obtained from a
molecular mechanics force field. The model is
then derived by minimizing the violations of all
the restraints. This can be achieved either by
distance geometry or realspace optimization.
9
1) Modeling by assembly of rigid bodies.
10
2) Modeling by segment matching or
coordinate reconstruction
11
3) Modeling by Satisfaction of Spatial Restraints
12
Loop Modeling
Loop modeling can be seen as a miniprotein
folding problem. loops are generally too short to
provide sufficient information about their local
fold. Segments of up to 9 residues are available
for loop modeling. Thus, the conformation of a
given segment is also influenced by the core stem
regions that span the loop and by the structure
of the rest of a protein that cradles the loop.
1. The ab initio loop prediction 2. Database
approach to loop prediction
13
Errors in comparatibe models
  • Errors in side-cahin packing.
  • Distortions and shifts in correctly aligned
    regions.
  • Errors in regions without a template.

14
d. Errors due to misalignments. e. Incorrect
templates.
15
Evaluation of Models
? Model Accuracy A model has the correct
fold? Sequence Identity, Energy based Z-score, or
conservation of the key functional or structural
residues
? Environments For instance, calcium-binding
proteins undergo large conformation changes when
bound to calcium
? Stereochemistry Bond lengths, angles, peptide
bond, side-chain ring planarity, chirality,
main-chain and side-chain torsion angles, and
clashes
? Spatial Features Packing, formation of a
hydrophobic core, solvent accessibility, spatial
distribution of charged groups, distribution of
atom-atom distances, atomic volumes, and
main-chain hydrogen bondings
16
Model Accuracy
17
Applications of Comparative Modeling
  • Design mutants to understand a proteins
    function
  • Identify active and binding sites
  • Identify, design, and improve ligands for a
    given binding sites
  • Model substrate specificity
  • Simulate protein-protein docking
  • Molecular replacement in X-ray structure
    determination
  • Refine models based on NMR constraints
  • Test and improve a sequence-structure alignment
  • Conform a remote structural relationship

18
(No Transcript)
19
Comparative Modeling in Structural Genomics
  • Large-scale comparative modeling
  • - Automation of all the steps in the modeling
    process is essential
  • (Fold assignment, template selection,
    target-template alignment, model generation, and
    model evaluation)
  • Providing many protein models from many genomes
  • Selection of a target protein for drug
    development.

Kolinskis Research Group
20
Conclusion
? Over the past few years, there has been a
gradual increase in both the accuracy of
comparative models and the fraction of protein
sequences that can be modeled with useful
accuracy.
? It is now possible to predict by comparative
modeling significant segments of approximately
one third of all known protein sequences. Low
accurate class 50 Medium accurate class 35
High accurate class 15
Write a Comment
User Comments (0)
About PowerShow.com