Transcript What is Protein Folding?

Blake Boling
Tamara Schneider
Proteins are a sequence of unbranched amino acids, which make the protein unique.
If folded wrongly, they do not function properly and can cause human
and animal diseases.
Protein folding is the process in which and amino acid sequence assumes a
functional three dimensional structure by folding and coiling. There are four
protein structures:
http://www.faseb.org/opar/protfold/fundstru.html
Misfolding of proteins can cause a variety of diseases:
Two possible tertiary structures:
Primary structure:
Formation of helices
and sheets
Secondary structure:
Actual folding
process
Tertiary structure:
optional
Quaternary structure:
 The amino acid sequence of a protein
 Codes for its Native confirmation
• Protein in α-helix structure
Two conformations of PrP (Prion):
• Protein in β-sheet structure
Left: normal PrPC
 Cystic Fibrosis
Right: proposed model of abnormal PrPSc
 Alzheimer’s Disease
 Consists of α-helices and β-sheets
 Necessary to obtain tertiary structure
Human:
 Creutzfeld-Jakob Disease (CJD)
 many cancers
 Native confirmation of the protein
 Held together by covalent bonds between
two cysteine residues and electrostatic
interactions
Animal:
 Mad Cow Disease
 Bovine Spongiform Encephalopathy
 Final assemblage
 Overall structure consisting of several
folded proteins
CJD and Mad Cow Disease are caused by the misfolding of Prion.
http://en.wikipedia.org/wiki/Prion
http://faq.distributed.net/bags/compute-cluster.png
Distributed Computing:
Query from a single computer using
resources from other computers and
providing the results.
Computational Techniques
Distributed Computing Projects
Distributed computing projects use computing resources banded together
throughout the world. Everyone can participate by downloading and
installing the corresponding software and providing the own unused CPU
cycles to the network.
Several computing techniques to simplify and study the folding of proteins have
been developed. Here are the most important examples, some utilize others.
The goal is to predict three-dimensional structure of proteins
from an amino acid sequence.
 Comparative Protein Modeling :
 Folding@home :
- By Stanford University’s Chemistry Department
- Second largest distributed computing project after
SETI@home (searching for extraterrestrial intelligence)
- Perform computationally intensive simulations of protein folding
- Aims to study the dynamics of protein folding
- Uses own networking infrastructure, but makes transition to BOINC
Computational Simulations of Model Proteins:
The Lattice Model
- Amino Acid sequence behave like single functional units
- Homology Modeling:
- Molecule folds to its least energy consuming state
• Based on the assumption that two homologous protein will share very
similar structures
• Given a solved amino acid structure from a homologous protein, it is
mutated into the unsolved structure
- Protein Threading:
- By Scripps Research Institute
• Scans the amino acid sequence of an unknown structure against
a database of solved structures
- perform computationally intensive simulations of protein folding
- Aims to specify what the tertiary structure of a protein will be
- Part of World Community Grid:
- In nature there are about 2000 distinct protein folding motifs that can be
used as templates
- Two approaches: Homology Modeling and Protein Threading
 Predictor@home:
 Human Proteome Folding Project:
- Uses previously solved structures as starting points
- Lattice proteins are highly simplified models of proteins
- Results are made publicly available
- Based on BOINC (Berkeley Open Infrastructure for Network Computing,
originally designed for SETI@home)
Protein Structure Prediction
http://www.bio.davidson.edu/Courses/Molbio/MolStudents/spring2003/Kogoy/protein.html
• A scoring function is used to assess the compatibility
Cartoon of a hypothetical crystalline lattice
 De Novo Structure Prediction:
- Tries to establish three-dimensional protein structure from scratch
• Largest public computing grid
- Requires vast computational resources
• Run by IBM
- Based on stochastic methods to search possible solutions
• Supports many different humanitarian projects
- Requires supercomputers or distributed computing