Under construction (throughout the workshop)!

Please also check the SHARP/autoSHARP wiki and the autoSHARP tutorial page there.



Here you will find hopefully useful information for the RapiData 2013 course at NSLS/BNL (21st - 26th April 2013). It should enable you to run autoPROC on data collected during the course as well as previously collected datasets you might have brought.

Make sure you have an autoPROC reference card available: autoproc_reference_card.pdf.

If you are new to Linux or the command-line, here are some maybe useful links for learning:

Local setup

On all machines used during the workshop you should be able to run

% process -h

giving you a help message for autoPROC. If this doesn't work please let us know!

Working on your own data

If you want to process data collected during the course or datasets you collected previously, make sure to know which beamline it was collected on: there are sometimes additional keywords necessary to describe the instrument correctly. There is a list of known beamlines on our autoPROC wiki; for the NSLS beamlines (apart from X25/Pilatus!) it seems necessary to run with

% process BeamCentreFrom=header:y,x ...

(where "..." is a shortcut for any additional arguments you might need).

The first requirement is to tell the program where the images are. There are several possibilities:

Running in the directory with images

This might be the easiest (if it fits your workflow and organisation). As an example:

% cd /where/ever/your/images
% process -d 01 | tee 01.lis

You could also make use of so-called symbolic links, eg:

% mkdir autoPROC
% cd autoPROC
% ln -s /where/ever/data/*.img .
% process -d 02 | tee 02.lis

Pointing to a directory with images

If all images for data collected on a crystal is in a single directory (and you prefer to run autoPROC in a different directory), just use

% process -I /where/ever/images -d 03 | tee 03.lis


% process -I ../Images -d 03 | tee 03.lis

Explicitely defining scans

Full control is available through the –Id flag. A command like

% process -Id "peak,/where/ever/images,Se_####.img,1,360" -d 04 | tee 04.lis


  • identifier
  • directory
  • filename template
  • first image
  • last image

Getting started

Remember to ensure you know which beamline data was collected on and if there are additional settings required. For X25/Pilatus images a good starting point would be

% process -d 05 | tee 05.lis

and for other NSLS beamlines probably something like

% process BeamCentreFrom=header:y,x -d 05 | tee 05.lis

(assuming that the beam centre value in the header is correct and only needs a simple transformation to work with autoPROC). If you expect problems during data-processing because the diffraction pattern already showed issues, maybe

% process BeamCentreFrom=header:y,x -M LowResOrTricky -d 06 | tee 06.lis

could be used. Make sure to read the NOTE and WARNING messages produced by autoPROC. There will also be a whole bunch of useful graphics created - which can be viewed eg with

% display 06/*.png

You should check your autoproc_reference_card.pdf and the online help available with

% process -h
% process -M list

And: just find us (in the X25 Blue Cubicle, during coffee breaks etc) and ask us anything you want to know!