Porting an EMAN1 refinement project to EMAN2.3
Note : This page contains a simplified version of the EMAN2.1 tutorial, and assumes you are familiar with EMAN1. The full tutorials provide a much better way of learning EMAN2. This page will get you started if you just want to rapidly switch a project from EMAN1. If you find yourself asking 'but what does THAT mean' when you read this, you're reading the wrong page!
A Quick (but worse) Alternative
Following the guide below will effectively start your project from an EMAN1 start.hed/img. There are a variety of advantages in doing things this way, as a number of things have improved in EMAN2 vs EMAN1. However, if your goal is just to get started quickly, even if the results are good, but perhaps not optimal, there is an alternative:
- create an empty folder adjacent to your EMAN1 refinement
- run e2import.py --import_eman1 --curdefocushint ../folder/start.hed
- note that the import script should work, but hasn't been updated in some time, so you will wind up with a few extra output files
run e2projectmanager.py and run the CTF -> CTF Autoprocess step (refer to the tutorial if you need details)
This will create a new EMAN2 project from start.hed/img. It will:
- split the particles by defocus
- undo the EMAN1 phase-flipping
- redetermine the CTF using the EMAN1 parameters as a starting point (you can use --curdefocusfix as an alternative)
- re-generate phase-flipped as well as filtered/shrunk particles
- build sets containing all of the particles
That's it, you're ready to start a refinement, though results may not be optimal.
While the above process should get you started quickly, to get optimal results, you should really follow the full single particle processing tutorial, starting from micrographs, with the following caveats:
- When importing micrographs, make sure you get the inversion correct. Particles in EMAN2 should be light on a darker background just like in EMAN1, and it is best if you start with the micrographs in this state.
- You can use the import .box file tool just like in the tutorial
EMAN2 is much more sensitive to having a sufficiently large [[EMAN2/BoxSize|box size]. Before doing the "Particles -> Generate Output" step, make sure the box size you plan to use is adequete.
- While there is no harm in doing unsupervised class-averaging, you probably already have a good 3-D starting model from your EMAN1 work, so you can likely skip the 2-D class averaging and initial model generation steps if you like