This version (2020/09/28 15:43) was approved by twagner.The Previously approved version (2020/09/25 10:29) is available.Diff

SPHIRE 1.3 / EMAN 2.31

We are happy to announce the release of SPHIRE 1.3. It introduces the new refinement strategy for filamentous samples, integrate our stand-alone deep learning programs and adds other helpful tools.

Helical processing with SPHIRE 1.3
Per particle CTF refinement. The image shows a refinement of appoferritin dataset with tilted stage. Each point represent a particle on the micrograph. The refined defocus values (left side) clearly shows the tilt axis. The right side is the a error estimate for every defocus value.
crYOLO can now directly used from SPHIRE 1.3
  • Hands free processing with AutoSPHIRE: While the individual steps can be configured and executed by hand, SPHIRE 1.3 includes AutoSPHIRE, an integrated tool that enables SPHIRE to run without the need for user supervision up until the creation of a high resolution reconstruction.
  • Per particle CTF refinement with error estimation. When you reach a resolution of about 4Å you can refine your CTF on a per-particle basis. This can improve the resolution by 0.3Å or more.
  • Integration of crYOLO 1.5 for particle picking. crYOLO is based on convolutional neural networks and can accurately pick about ten micrographs per second.
  • Integration of automatic 2D class selection with Cinderella. This tool is based on a deep learning network to separate 2D classes from .hdf / .mrcs files into good and bad classes. It uses the same deep neural network as crYOLO and comes pre-trained on a set of good and bad classes. Cinderella was written to automate cryo-EM data processing.
  • The new compare re-projections tool can now be used to produce side-by-side comparisons of the 2D class averages with the respective projections of a 3D model.
  • Visualization of 2D power spectra in CTER.
  • Minor bug fixes.

The installer uses Anaconda, which should install independently of the choice of Linux distribution. The Mac installer is experimental.

Here are the downloads for Mac and Linux:

SPHIRE does not support Windows

These external tools will be needed to run all tools in SPHIRE:

TranSPHIRE for automatated preprocessing

Furthermore, we recommend to use TranSPHIRE for automated preprocessing: TranSPHIRE GitHub page

  • downloads/sphire_1_3.txt
  • Last modified: 2020/09/28 15:43
  • by twagner