Software

SPIDER (System for Processing Image Data from Electron microscopy and Related fields) is an image processing system for electron microscopy, developed since 1978 by Joachim Frank and his group. SPIDER is written in FORTRAN and is used for mathematical manipulation of images and their contents, with special emphasis of operations required for three-dimensional electron microscopy.

Contributors over the past 30 years include: J. Frank, B. Shimkin, H. Dowse, L. Miranda, C. Mannella, J. P. Bretaudiere, A. Verschoor, M. Radermacher, A. Leith, J. M. Carazo, P. Penczek, L. Odesanya, Y. H. Li, M. Ladjadj, Y. Chen, K. R. Lata, J. Zhu, W. P. Liu, B. Rath, C. Yang, B. Baxter, R. Hegerl, A. Frangakis, T. Shaikh, J. LeBarron, and N. Boisset.

For a recent paper on the use of SPIDER in the 3D reconstruction of macromolecules, see [Shaikh et al., 2008] in the publication list. SPIDER and the associated packages WEB, JWEB, and SPIRE are distributed freely, and are still maintained at the Wadsworth Center by RVBC, the NIH/NCRR-supported Resource for the Visualization of Biological Complexity. Please visit the official SPIDER website for more information. Here are some quick links:

• batches tar.gz file
• List of operations
• Techniques
• 3D reconstruction
• Euler angle

Xmipp, “X-Window-based Microscopy Image Processing Package”, (J. Struct. Biol. 148(2), 194-204) is a suite of image processing programs, primarily aimed at single-particle 3D electron microscopy. A version is installed on our cluster.

External links:

• Xmipp home page (Wiki page)
• Ancient home page

UCSF Chimera is a highly extensible program for interactive visualization and analysis of molecular structures and related data, including density maps, supramolecular assemblies, sequence alignments, docking results, trajectories, and conformational ensembles. High-quality images and animations can be generated. Chimera includes complete documentation and several tutorials, and can be downloaded free of charge for academic, government, non-profit, and personal use. Chimera is developed by the Resource for Biocomputing, Visualization, and Informatics and funded by the NIH National Center for Research Resources.

External links:

• UCSF Chimera home page
• Image gallery
• Animation gallery
• User’s guide: everything to get started with Chimera
• Tutorials:
  • Getting Started
  • Tutorials Index
  • More tutorials
  • Experimental features

Conversion of angles from SPIDER to Chimera

SPIDER

; refangles.dat

Chimera -> command line
reset ;; reset the viewing angle to 0 0 0
turn z -1*phi
turn y -1*theta
turn x 180
(The order matters.)
Then we get the projection view same as SPIDER -> Alignment/projs/

HIGHLIGHT: Making movies in Chimera
There are several methods with Chimera; first read this guide.

• Graphical interface: the movie recorder, cf
• Script method: The “movie” command
• MD Movie: for tool for trajectory (or ensemble) viewing and analysis.
• Morph Map: morphs between two related sets of volume data (maps).

With the movie command, you will write all the instructions in a text file, that you will source in Chimera from the command line. This is a very reproducible method, that allows you to reuse simple scripts for a nice effect. You first need to be familiar with the basic commands and command-line selections.

Example: rotating the glutamate synthase. This is a simple animation in which a cryoEM volumes rotates, and fades, showing the underlying structures. The command file can be downloaded here.

How to… launch a refinement on the cluster

• Connect with ssh -X username@156.111.6.184
• You can check the disk space with “df -h”
• You can check the space usage of your files with “du -sh”
• From another terminal on your machine, transfer your files onto the master node:
  scp –r your_files_or_directories 156.111.6.184:destination_path_on_the_master_node
• Check you have the input files, check your parameters in refine_settings.pam
• Check that subscribe.perl is running (with a ps command) (ps -ef | grep subscribe)
• If not, start it by typing “startsub.perl”
• Copy a local version of the SPIDER binary in your working directory
• Then you are ready to launch your refinement:
  ./spider pam/ext @pub_refine 0
  OR
  ./spider pam/ext @pub_refine 0 & to send the job in background (recommended in case your terminal crashes…)
• Monitor the resolutions at the end of each cycle in final/resolutions.ext

How to… launch a non spider job on the nodes of the cluster using pubsub

Publish has been written to distribute SPIDER-only jobs on the nodes of the cluster. But there is a way to “trick” it to deal with other jobs such as shell scripts. You need for that to write a “dummy” SPIDER job that will be recognized by publish and that will launch your job.

Example: Analyse the output of ctftilt (ctftilt.log) and calcultate the defocus of each particle.
Since this takes a looooong time, this is worth using the cluster, with defocuses_from_ctftilt_cluster.spi, calling defocuses_from_ctftilt.sh and using a “dummy” spider script to pilot the execution of the shell script on the nodes: dummy.spi.