QUEEN

From NMR Wiki

Jump to: navigation, search

QUEEN stands for: "QUantitative Evaluation of Experimental Nmr restraints"

QUEEN objectively sorts distance and dihedral angle restraints by "importance" without ever calculating 3D structures.

Usually only ~10% of restraints contain most information about the structure - therefore these restraints have to be correct for good structure calculations.

Project home page is here

Contents

Summary

The QUEEN program estimates information content in each restraint by calculating how much a given restraint improves uncertainty in the distance geometry matrix (DG matrix). DG matrix is calculated by NIH-XPLOR which is called internally from the QUEEN package.

Knowing which restraints are more important allows to save a lot of time spent on checking correctness of restraint assignments (since it is known that e.g. typically only ~10% of NOE restraints provide most information for the structure calculations).

Limitations: (1) program only works with proteins; does not work for DNA, RNA or other molecules; (2) only 32-bit linux version available

Running time: depends on number of restraints in the table and size of polypeptide, typically takes from half hour to 1 day on a relatively fast desktop computer.

Requirements: python, NIH-XPLOR, 32-bit Linux OS

Quick guide

Preparing input

  • go to directory where queen is installed
  • come up with a name for your project
  • type
./generate.py --project <projectname>
  • this will create the necessary directory structure
  • collect restraint files in your project directory
  • check restraint files (restraint files must have .tbl extensions and must be in XPLOR format)
./restraints.py --check distance-restraints.tbl -o good-distance-restraints.tbl
./restraints.py --check -d dihedral-restraints.tbl -o good-dihedral-restraints.tbl
  • the good ones will be saved into the files specified with -o flag
  • copy good restraints to <projectname>/data/restraints directory
  • set up data set file <setname.list> (extension is important), place file into <projectname>/data/datasets
  • almost done. Now you'll need sequence, .psf and .pdb files. You have two choices

Starting with a sample pdb file (easier setup)

  • type
./generate.py --pdb2all <projectname> sample.pdb 
  • this will create the psf and protein sequence files

No pdb file (harder setup)

  • prepare a sequence file in three letter format (explained below)
  • if you have one-letter sequence file with a single chain, this script will do the job
  • type
./generate.py --seq2psf <projectname> <sequence file>
  • this should create psf file
  • now type
./generate.py --psf2tem <projectname> <projectname>/data/sequence/protein.psf
  • this will create template pdb file from the .psf file

Checking input and running the program

  • check the project
queen.py --check <projectname> <setname>
  • run it
queen -Iuni <projectname> <setname>

Sequence file format

>A (these chain names must be present)
ALA
GLU
PRO
...
>B
...

Format for dataset files

dataset files must have .list extensions and must be placed into <projectname>/data/datasets directory


NAME = NOE restraints
TYPE = DIST
FILE = good-noe-restraints
//
NAME = Dihedral angle restraints
TYPE = DIHE
FILE = good-dihedral-restraints
INFO = no
//

INFO field is optional. If "INFO = no" is used, these restraints won't be added to the ranked output

References

  1. Nabuurs, SB and Spronk, CAEM and Krieger, E and Maassen, H and Vriend, G and Vuister, GW. Quantitative Evaluation of Experimental NMR Restraints. J Biomol NMR 3:55--66, 1993. BibTeX [queen]
Personal tools