Transcript Flexible Multi-scale Fitting of Atomic Structures into Low

Flexible Multi-scale Fitting of Atomic Structures into Lowresolution Electron Density Maps with Elastic Network
Normal Mode Analysis
Tama, Miyashita, Brooks. 2004. 337. 985-999
Used NMA to fit high resolution xtal structures to low resolution EM
maps. Mostly concerned with conformational change.
Also employed flexible fitting method with Calpha only claiming this
allowed larger systems to be computationally tractable.
Show that this method works better than Situs package on the
targets they tested.
ProMate: A Structure Based Prediction Program to
Identify the Localization of Protein-Protein Binding Sites
Neuvirth, Raz, Schreiber. 2004. 338. 181-199
Binding surfaces share common properties that distinguish them
from non-binding surfaces
Interesting review of binding surface properties, as compiled
through many investigations
I think they draw some questionable conclusions.
The algorithm correctly predicts the binding site in about 50% of
cases, random would be 13%.
Clear preference for binding interface to exist in region of loops.
Mean loop length 11.17
Testing a Flexible-receptor Docking Algorithm in a
Model Binding Site
Wei, Weaver, Ferrari, Matthews, Shoichet. 2004. 337. 1161-1182
Flexible receptor obtained by ensemble of xtal structures
Linear scaling opposed to exponential scaling
Recommend reading, wont say much more.
Can Contact Potentials Reliably Predict Stability
of Proteins?
Khatun, Khare, Dokholyan. 2004. 336. 1223-1238
Contact potential describes the relative free energy of a given
protein conformation by evaluating the interaction among residues
Evaluate a set of mutants and try to predict the ΔΔG for a single
amino acid mutation
- contact potentials parameterized from experimental data
(experimental error ~ kt)
They show that contact potentials are neither accurate nor
transferable and claim that it is impossible to develop a set of
transferable contact parameters to assess changes in stability of
multiple proteins.
Protein Flexibility in Ligand Docking and Virtual
Screening to Protein Kinases
Cavasotto, Abagyan. 2004. 337. 209-225
ICM-flexible receptor docking algorithm (IFREDA)
-generates a discrete set of receptor conformations
Claims:
-both side chain rearrangements and backbone
movements, including loop movements are taken into
account
-this method helps when no holo structure available, they
dock ligand into pocket, rotate and translate, reoptimize
pocket, use this ensemble for docking