Issues with 2D Classification of Negative Stain data

Environment:
Current Version - v2.9.0
CUDA Version - 10.1 with correctly specified path
OS - Ubuntu 18.04.3

Issue
I am trying to perform 2D classification of NS data and with two of my 5 data sets, I cannot get the particle picking or 2D classification to work properly.

Here is a data set that works well then I will show what is going awry with the other two.

Since I am a new user, apparently I can only upload 1 image, so here is everything in one image. Im sorry.

Does anyone have an idea of what I can do to get this to work correctly?

Flip your images with e2proc2d.py before importing them, then treat them like cryo images (dark particles).

Here’s more detail for my approach: Easy template picking

Also you may need to change your CTF parameters. I would set the low resolution to something lower - so that the first peak isn’t masked, say 50 A. The amplitude contrast should be ~0.4. I recommend using CTFFIND4 over gCTF, and then always using the “fast” mode.

Good luck!

1 Like