Project

General

Profile

Actions

Bug #4122

closed

ctf plot for envelope normalization looks wrong on an unusually good image

Added by Anchi Cheng about 8 years ago. Updated over 7 years ago.

Status:
Closed
Priority:
Urgent
Assignee:
Category:
-
Target version:
Start date:
04/19/2016
Due date:
% Done:

0%

Estimated time:
Affected Version:
Appion/Leginon 3.3
Show in known bugs:
Workaround:

Description

The image attached is taken on lacey carbon on Krios.
300 kV, Cs=2.7 mm pixel size 1.09699 Angstrom

I fit it to 2.2 Angstrom and modify apDisplay to display 1D plot to 2.3 Å.

The result of the normalization looks good, though.


Files

Actions #1

Updated by Neil Voss about 8 years ago

I will take a look, the bad envelope comes from a bad noise fit.

Actions #2

Updated by Neil Voss about 8 years ago

  • Project changed from TiltPicker to Appion
  • Affected Version set to Appion/Leginon 2.2.0
Actions #3

Updated by Anchi Cheng almost 8 years ago

  • Priority changed from Normal to Urgent
  • Affected Version changed from Appion/Leginon 2.2.0 to Appion/Leginon 3.3

Neil, Please work on this asap. I really want to merge your ctf stuff to myami-beta but I can not do so until it works right. It might be as simple as a problem of plot scale. As you can see, and I observed on other plots that the minimum in Y is 1.0, not 0.0 which I assume it should be.

Actions #4

Updated by Neil Voss almost 8 years ago

Sorry, I will get on it right now.

Actions #5

Updated by Neil Voss almost 8 years ago

Works now, I had to adjust the parameters which floors the data.

Actions #6

Updated by Neil Voss almost 8 years ago

  • Status changed from Assigned to In Test
  • Assignee changed from Neil Voss to Anchi Cheng
Actions #7

Updated by Anchi Cheng over 7 years ago

Neil,

3d4e2a6b further changes this. I have many images from K2 camera that end up with flat line from the last commit. This new change at least make it a bit better. These images have normexprotdata maximum smaller or close to 1.0.

The result of this commit still gives the first section envelop (magenta) line away from the data (blue) and fit(green) in the example attached. Therefore, I might still missed something. Hope you can find out where.

For other testing, try the appion testdata set of GroEL images. I can also give you more images of this type or phase plate images which are usually -0.5 um defocus and with large values at low resolution.

An example is attached.
ctffind4 result

# Output from CTFFind version 4.1.5, run on 2016-10-05 15:19:28
# Input file: 16oct04c_a_050gr_012sq_06hl_04ex-a.mrc ; Number of micrographs: 1
# Pixel size: 0.900 Angstroms ; acceleration voltage: 300.0 keV ; spherical aber
ration: 2.20 mm ; amplitude contrast: 0.07
# Box size: 1024 pixels ; min. res.: 50.0 Angstroms ; max. res.: 3.0 Angstroms ;
 min. def.: 1000.0 um; max. def. 46929.8 um
# Columns: #1 - micrograph number; #2 - defocus 1 [Angstroms]; #3 - defocus 2; #
4 - azimuth of astigmatism; #5 - additional phase shift [radians]; #6 - cross co
rrelation; #7 - spacing (in Angstroms) up to which CTF rings were fit successful
ly
1.000000 22435.417969 21929.798828 73.746925 0.000000 0.019749 6.433184

Appion input

ctffind4.py   --ampcontrast=0.07 --fieldsize=1024 --resmin=50 --resmax=3 --defstep=0.1   --numstep=25 --dast=0.05 --best-images --min_phase_shift=10   --max_phase_shift=170 --phase_search_step=10 --runname=ctffind4run4   --rundir=/nis/ucsd-02/appion/16oct04c/ctf/ctffind4run4 --preset=ex-a   --commit --projectid=4 --session=16oct04c --no-rejects --no-continue   --expid=19 --jobtype=ctfestimate --no-wai

Actions #8

Updated by Neil Voss over 7 years ago

  • Status changed from In Test to Closed

these changes look good.

Actions

Also available in: Atom PDF