Transcript What is the difference between RMSF? RMSD? B-Factor?

Comparing Data from MD simulations
and X-ray Crystallography
•
•
•
•
What can we compare?
3D shapes (RMSD)
Atomic motions (B-value, RMSF)
Solvent accessibilities (SASA)
Tools for Analyzing Protein 3D Structures
AMIGOS - Reads an RNA PDB file and outputs a complete table of torsion angle calculations.
APBS - Software for evaluating the electrostatic properties of nanoscale biomolecular systems.
Atlas of Protein Side-Chain Interactions - depicts how amino acid side-chains pack.
Bioinformatics Toolkit - a platform that integrates a great variety of tools for protein sequence
analysis - Max-Planck Institute for Developmental Biology.
CATH - Hierarchical classification of protein domain structures.
CAVENV - Designed to help visualise cavities in a protein structure.
CAVER - Automated calculation of pathways leading from buried cavities to outside solvent in
static and dynamic protein structures.
CaPTURE - Identify energetically significant cation-pi interactions within proteins.
CASTp - Computed atlas of surface topography of proteins.
ConSurf - Identification of Functional Regions in Proteins.
COPS -(Classification Of Protein Structures) web server which is a next-generation web
application that provides fast and intuitive access to the entire set of currently available
protein structures. COPS organizes structural domains by quantified structural similarities
which can be visualized immediately with Jmol. Additionally, structural biologists can employ
iCOPS to classify their own structures.
DDMP -- Differences Distance Matrix Program.
Dynamite - Makes predictions about the way a protein moves given only a PDB file.
Dynapocket - Prediction of likely configurations of the atoms of binding pocket.
DynDom - Determine domains, hinge axes and hinge bending residues in proteins.
dssp2pdb - Converts dssp output to pdb.
eF-site - Electrostatic surface of functional site.
elNemo - Compute the low frequency normal modes of a protein.
Entangle - Creates a listing of interactions that occur at the interface between the nucleicacid/protein complex. These interactions are then classified into hydrogen bonds,
electrostatic, hydrophobic, and Van der Waals.
ESCET - Analyse and compare three-dimensional protein structures.
extpdb - If you cut up pdb files, this is the utility for you.
FISH - Domain identification server.
GETAREA - Calculation of solvent accessible surface areas, atomic solvation energies and their
gradients for macromolecules.
HBexplore - Geometrical analysis of hydrogen bonds in biological macromolecules.
InterPro - Database of protein families, domains, repeats and sites.
iMolTalk - Structural Bioinformatics Toolkit; an interactive, Internet-based service for
computational analyses in Structural Biology.
MACiE - contains 223 fully annotated enzyme reaction mechanisms (Janet Thornton)
MOLEOline - Web-based tool to find/analyze molecular channels, tunnels and pores.
MolProbity - An expert-system consultation about the accuracy of a macromolecular
structure.
MOTIF - Sequence motif search from GenomeNet.
MSDMotif - Provides information about ligands, sequence and structure motifs.
NACCESS - Calculates the accessible area of a molecule from a PDB format file.
NCBI BLAST - Finds regions of similarity between biological sequences.
NQ-Flipper - Validates and corrects asparagine and glutamine side-chain amide rotamers in
protein structures. An interactive Jmol applet highlights the most offending interactions
between atom pairs.
OCA - A browser-database for protein structure/function.
OnD-CRF - Order and Disorder in proteins prediction server.
Parvati - Protein Anisotropic Refinement Validation and Analysis Tool
PDBsum - At-a-glance overview of every macromolecular structure deposited in the PDB.
PDB Report Database - Reports describing structural problems in PDB entries.
PDB Structure Quality - Determine if your structure is on-par with others in the PDB.
PDB2PQR Server - Enables a user to convert PDB files into PQR files.
Pfam - Large collection of protein families.
PIqsI - investigation and curation of quaternary structures.
PISA - An interactive tool for the exploration of protein Interfaces, Surfaces and
Assemblies.
Pre-BI - Prediction of biological interfaces.
PredictProtein - A service for sequence analysis, structure and function prediction.
Probe - Evaluate atomic packing, either within or between molecules. It generates “contact
dots” where atoms are in close contact.
ProDom - Protein domain families.
PRODRG-2 Server - Convert small molecules in PDB format to various topology formats.
ProFunc - Prediction of protein function from 3D structure; EBI services.
ProMate - Predicts the location of potential protein-protein binding sites.
PROPKA - Empirical Prediction and Interpretation of Protein pKa Values.
Protein Calculator - Protein properties calculator.
Protein Structure Prediction - Molecular modelling site.
Proteopedia - The collaborative, 3D encyclopedia of proteins and other molecules. Article
ProtSkin - Converts a protein sequence alignment to a property file used to map the
sequence conservation onto the structure of a protein.
PROTORP - A tool to analyse the properties of interfaces of protein-protein associations.
PSAIA - Software tool that integrates several algorithms for protein interactions and structure
geometry analysis of protein complexes into a single application.
RAPPER - An ab initio conformational search algorithm for restraint-based protein modelling.
ROBETTA - Full-chain protein structure prediction server.
SCOP - Structural classification of proteins.
SFCheck - Assessment of X-ray data and/or agreement between atomic model and X-ray data.
SPASM Server - Recognition of spatial motifs in protein structures.
SSBOND - predicts sites where disulphide bonds can be introduced.
STAN - Structure analysis server.
StrucTools - Common structural biology calculations with PDB coordinates.
SURFNET - Generates surfaces and void regions between surfaces.
TLSMD - Inference of TLS motion from single crystal structures.
TOPS - Compare a structure (PDB Format) to a database of structures.
Web-Tools - Web-Accessible Scientific Applications (scientific supercomputing at NIH).
WHAT IF - Homology modelling, drug docking, electrostatics, validation and visualisation.
3Dee - Database of Protein Domain Definitions.
3DNA - Analysis, rebuilding, and visualization of three-dimensional nucleic acid structures.
What is the difference between
RMSF? RMSD? B-Factor?
Read the book!
To monitor changes in 3D structure,
RMSD is often plotted as a function of MD simulation time
When does the simulation become
stable?
How big a difference is there between
the stable structure and the reference
structure?
How can RMSD be used to tell when a
simulation has reached structural
equilibration?
The Template Modeling Score or TM-score is a measure of similarity
between two protein structures with different tertiary structures. The TMscore is intended as a more accurate measure of the quality of full-length
protein structures than the often used RMSD. The TM-score indicates the
difference between two structures by a score between (0,1], where 1
indicates a perfect match between two structures[1]. Generally scores
below 0.20 corresponds to randomly chosen unrelated proteins whereas
structures with a score higher than 0.5 assume roughly the same fold
What is solvent accessible surface area
and how is it calculated?
How is solvent accessibility calculated?
What is the difference
between the van der
Waals surface and the
solvent accessible
surface?
SASA values are useful for
identifying the degree to which
each amino acid in a protein is
exposed or buried in the core
Comparing different structures of the
same protein helps identify regions that
are more flexible – they will be a little
different in each structure
Comparison between multiple
crystal structures identifies flexible
regions
Comparison between crystal
structures and MD data confirms
flexible regions
Why does the MD data show
larger fluctuations (B-factors)?
Can compare theoretical data (MD
structures) to crystal data such as
Ramachandran plots
How can we get a numerical value that
gives a measure of the “goodness” of a
model? “Z-Score”
Z-scores
For any data set that can be described by a normal
distribution, the distribution is perfectly symmetric
about 0.
A Z-score compares a single data point to the normal
distribution. The Z-score for a value is the number of
standard deviations that value is away from the mean
of the distribution
Z-score = (x – mean)/stdev
Example: Find the z-score
corresponding to a raw score of 132
from a normal distribution with mean
100 and standard deviation 15.
Solution
Z-score = (x – mean)/stdev
= (132 – 100)/15
= 2.13
A Z-score can be positive or negative and
gives a quantitative idea of how far a given
value is from the average for similar data.
Nucleic Acids Res. 2007 Jul;35(Web Server issue):W375-83. 2007 Apr 22.
MolProbity: all-atom contacts and structure validation for proteins and
nucleic acids.
Davis IW, Leaver-Fay A, Chen VB, Block JN, Kapral GJ, Wang X, Murray LW,
Arendall WB 3rd, Snoeyink J, Richardson JS, Richardson DC