8. Protein, Structure, PDB Flashcards
AA built up
N-teminus-> peprtide bond -> C-terminus
AA end Carboxyl terminal end
free rotation between
C-C, phi and psi angles, cannot be any angle due to steric hindrance
Ramachandran plot
plot of phi and psi
visualizes allowed angle combinations
alfa helix
optimal use of hydrogen bonds
3,6 residues/turn
right handed most often
beta strand
forms beta sheets
backbone H-bonding between strands
side chains are away from backbone
coil&loops
- lot of variability
- frequently part of active and binding sites
- open/close protein
general rules
- bury hydrophobic residues in the interior
- contribute to stability
- a-helices and b-sheets are generally clustered
- segments close in sequence tent to be close in space
Motif
super-secondary structure
- simple combinations of few secondary structure elements with a specific geometric arrangement
- can be associated with specific function
Methods for structure determination
- X-ray
- NMR
- EM
x-ray
+ atomic resultion
+ high accuracy
- requires crystals
- much protein needed
NMR
+ Info on dynamics
+ less protein needed
- requires labelling
- water’s not resolved
EM
+large proteins
+ little protein needed
-low resolution
-difficult sample prep
PDB
- get coordinate files
- database contain x-ray, NMR, EM
- 4 letter/number code
Fold classification
similar fold -> similar function
- SCOP, CATH
- Compare folds, domain
- class, fold, superfamily, family
- based on PDB
3D comparisons
- key approach to protein structure analysis
- detect distant homologs
- structure classification
- requires superposition
- RMSD (root mean square deviation)
commonly used structure comparison tool
DALI
- based on intramolecular distances
- used information from PDB
key techonologies for protein ident.
2D gel technology
Staining/scanning
MS
Databases
Time-of-flight
velocity of the ion depend on the mass-to-charge ratio
Gyros
A type of mass spec, but you have an IMAC-CD with channels
- Measure speed and charge
Database searching
From eg MS you get a fingerprint which you can put in a database to identify protein
-Thus this does not work on unknown proteins
KEGG
Pathway database
- Today pharmaceutical companies are adding their drugs to targets in this database
SFLO
structure - function linkage database
- connect sevaral databases such as BLAST, PDB, PSI-BLAST, KEGG
