Data Reduction
Structural Molecular Biology Laboratory, ChemM230D
A one degree oscillation photograph
Suggested Reading Materials

1)Automatic indexing of rotation diffraction patterns by Wolfgang Kabsch, Journal of Applied Crystallography (1988), vol. 21, 67-72.

2)Processing of X-Ray Diffraction Data Collected in Oscillation Mode by Zbyszek Otwinowski and Wladek Minor, Methods in Enzymology, 276, 307-326.

3)XDS/XSCALE manual, XDSwiki

4) Solvent Content of Protein Crystals by B.W. Matthews, J. Mol. Biol. 33, 491-497 (1968).

5) Pre-lecture presentation Please read this outline before class.

6) Powerpoint presentation presented in class.

7) Zoom recording of presentation from Jan 12, 2021.

 
Assignment & Procedures
 
Assignment: Table of Data Processing Statistics 
will be completed in lab
Illustrations
Objective: To produce a crystallographic data statistics table typically found in structural biology journals. 

Method: Copy the format of the table on the left.  Substitute the data from your Scalepack log files for the data in the table.  Report data statistics on the data set that you collected (native,  derivative, or PMSF).   Submit your Table 1 to Mike or Duilio when completed .
 
A typical Table 1  for an MIR experiment, adapted from Blaszczyk et al., Crystallographic and Modeling Studies of RNase III Suggest a Mechanism for Double-Stranded RNA Cleavage, Structure, Vol. 9, 1225-1236, December 2001.
 
Part One:
Autoindexing & Integration
Illustrations
Objective: Define the crystal orientation, lattice and unit cell parameters for the data set you collected last week.  We have 360 data images, each image containing hundreds of reflections. In order integrate the intensities of the reflections we must be able to fit the geometric position of each reflection to a point in the reciprocal lattice. Hence, each reflection is indexed with a specific HKL value. The initial fitting is typically performed using a single data image and requires the knowledge of several data parameters. Somer are known a priori, some are not. The known parameters are input in the autoindexing script. The unknown parameters are output for use in the next step- integration. Pay close attention to the location of each of these parameters in the autoindexing scripts.

KNOWN PARAMETERS
1) wavelength
2) crystal to detector distance
3) position of the direct beam (this defines the origin of the reciprocal lattice.)
4) oscillation angle
5) list of 14 possible Bravais lattices
6) list of the geometrical positions of reflections on a single image


UNKNOWN PARAMETERS
7) unit cell parametsrs (3 lengths, 3 angles)
8) crystal orientation (3 angles)

Procedures
1) Go to your working directory.  Display the first image of your data set.  Run the autoindexing script.

2) Choose which Bravais lattice with the highest symmetry consistent with the observed image(distortion percentage between 0-2%).

3) new space group p43212, fit all  @gogo.

4) Judge the goodness of fit of the predicted lattice to the diffraction pattern (visually, and with chi**2 statistics).  If the fit is good, then most spots should be indexed. 

5) Adjust the mosaicity, background and spot size.

6) List parameters, cut and paste into integration script file.  Begin integration for all frames collected

The 14 Bravais lattices
 
Part Two:
ScalePack
Illustrations
Objective: To scale together symmetry related intensity measurements and verify systematic abscences. 

Idea: Just as there is an asymmetric unit in the unit cell of the crystal, there is an asymmetric unit in the reciprocal lattice.  It is the smallest group of reciprocal lattice points that can reproduce the entire reciprocal lattice by symmetry operations.  Evaluating the statistics from scaling can help you determine whether you have chosen the correct symmetry operators (space group) for the crystal.

Statistics:
 Measured reflections- the total number of  reflection observations in your data set.  In this example it is 568,530.  The more the better.
Unique reflections-the total number of reflections after symmetry averaging.  This number is a subset of the total number of repcrocal lattice points in the asymmetric unit. If you take the number of Measured reflections divided by unique reflections you get the overall redundancy of the data set.  Bigger unit cells have more unique reflections.
Completeness overall- The percent completenes of your data set.  Measures whether you collected all the reflections in the asymmetric unit. The importance of completeness is illustrated in this movie from James Holton.
Completeness in the last shell-The percent completeness of the highest resolution shell of your data set.  You can't claim you have 1.8 Angstrom data unless this shell is fairly complete.  This nmber keeps you honest.
Rsym overall- Measures the agreement of symmetry related observations of a reflection.  In this example, the symmetry related reflections agree to within 5.6%.
Rsym in the last shell- Measureds the agreement of symmetry related observations in the highest resolution shell.  Don't accept shells above 40%.
I/sigma-A measure of the signal to noise ratio.
 Resolution- of the data set is determined by a combination of statistics pertaining to the last shell (high resolution shell) of the data set. My personal criteria are that the I/sigma of the highest resolution shell be greater than 2.0 and the Rsym be less than 50%. If the highest resolution shell does not meet these criteria, then this shell should be discarded and the next highest shell that meet these criteria should be reported as "the high resolution shell". The importance of completeness is illustrated in this movie from James Holton.

Look for systematic absences at the bottom of the log file.  Verify that they are absent.
 
Part Three:
Scalepack2mtz
Illustrations
Objective: To prepare data for the Patterson calculation. 

Idea: Next week we will be calculating difference Patterson maps using the CCP4 crystallographic suite of programs.  CCP4 requirs a specific format for the data set called mtz.

Procedures:
Make a ccp4 sub directory.  Go to that directory and type ccp4i.  Go to the programs list window.  Select scalepack2mtz.  Fill in the boxes as shown in the illustration on the right.

Instructor's preparations

Back to CHEM M230D course syllabus 

[Overview] ·[Facilities] · [People] · [Services] ·[Lectures] · [BioLinks] · [Stats] ·[Search]