how to run frealign without doing CTF correction?

Submitted by xzhang2005 on Tue, 2014-09-02 21:50

Forums

Frealign

Hi,

In two cases, I need to run frealign without doing CTF correction on the input images: (1) images collected using a phase plate; (2) regular defocused images, but with phase flipped. How can I setup frealign to run such data? Thanks.

Bests,
Xing

niko Wed, 2014-09-03 11:51

You can switch off CTF correction by setting the amplitude contrast to -1.0.

xzhang2005 Sun, 2014-09-21 20:11

It doesn't work by setting WGH=-1.0. I set all defocus=0.0 in the input file, and the following is the output from frealign_v8:

=====================================
C Date and time 21-09-2014, 17:06 Frealign V8.11 - 26.08.13
C Image format . . . . . . . . . . . . M
C Mode . . . . . . . . . . . . . . . . 3
C MASK for parameter refinement. . . . 1 1 1 1 1
C Magnification refinement . . . . . . F
C Defocus refinement . . . . . . . . . F
C Astigmatism refinement . . . . . . . T
C Defocus ref. of individual particles F
C Ewald sphere correction. . . . . . . 0
C Beautify the final real space map. . T
C Apply Wiener filter to final map . . T
C B-factor correction of final map . . F
C Write out matching projections . . . F
C Calculate FSPR and FSC curves. . . . 0
C Calculate more statistics. . . . . . F
C Padding factor for reference volume. 1
C Outer Radius of object [Angstroms] . 350.00
C Inner Radius of object [Angstroms] . 250.00
C Pixel size [Angstroms] . . . . . . . 4.48000
C % Amplitude contrast . . . . . . . . -1.00
C STD level for 3D mask. . . . . . . . 0.00
C Phase res. / B factor constant . . . 5.00
C Average phase residual for weighting 75.00
C Symmetry card as input . . . . . . . I2
C Number of symmetry operators . . . . 4
C 1 0.8090 0.3090 0.5000
C 1 0.3090 0.5000 -0.8090
C 1 -0.5000 0.8090 0.3090
C 2 0.0000 0.0000 1.0000
C 2 1.0000 0.0000 0.0000
C 2 0.0000 1.0000 0.0000
C 3 1.0000 0.0000 0.0000
C 3 0.0000 -1.0000 0.0000
C 3 0.0000 0.0000 -1.0000
C 4 -1.0000 0.0000 0.0000
C 4 0.0000 1.0000 0.0000
C 4 0.0000 0.0000 -1.0000
C Angular step size for search . . . . 4.0
C First particle . . . . . . . . . . . 1
C Last particle. . . . . . . . . . . . 10
C Relative magnification . . . . . . . 1.0000
C Densitometer step size (microns) . . 56.0
C Phase residual target. . . . . . . . 10.00
C Phase residual threshold . . . . . . 90.00
C Cs [mm]. . . . . . . . . . . . . . . 2.70
C Voltage [kV] . . . . . . . . . . . . 300.00
C Beam tilt Tx, Ty [mrad]. . . . . . . 0.00 0.00
C Resolution of reconstruction . . . . 10.000
C Low resol. limit refinement. . . . . 100.000
C High resol. limit refinement . . . . 20.000
C Defocus uncertainty. . . . . . . . . 0.000
C B-factor for parameter refinement. . 0.000
C Input image stack /datadisk/frames/xingzhang/BPP_phasePlate/frealign
C Input parameter file /datadisk/frames/xingzhang/BPP_phasePlate/frealign
C Output parameter file refine.par_1_10
C Output shifts file refine.shft_1_10
C Search angle incr. 3.91 No. of search angles: 45
C 3D reconstruction file /datadisk/frames/xingzhang/BPP_phasePlate/frealign
C 3D weights file refine_weights_1_10
C 3D reconstruction halfset 1 refine_qfactor_1_10
C 3D reconstruction halfset 2 refine_amplitudes_1_10
C 3D ave phase residual file refine_phasediffs_1_10
C 3D point spread function refine_pointspread_1_10
C
C PSI THETA PHI SHX SHY MAG FILM DF1 DF2 ANGAST PRESA DPRES
1 0.00 0.00 0.00 0.16 0.07 125000. 10001 0.0 0.0 0.00 0.00 -89.99
2 0.00 0.00 0.00 0.31 0.75 125000. 10001 0.0 0.0 0.00 0.00 -89.99
3 0.00 0.00 0.00 0.41 0.57 125000. 10001 0.0 0.0 0.00 0.00 -89.99
4 0.00 0.00 0.00 0.29 0.45 125000. 10001 0.0 0.0 0.00 0.00 -89.99
5 0.00 0.00 0.00 0.06 0.32 125000. 10001 0.0 0.0 0.00 0.00 -89.99
6 0.00 0.00 0.00 0.14 0.37 125000. 10001 0.0 0.0 0.00 0.00 -89.99
7 0.00 0.00 0.00 0.80 0.50 125000. 10001 0.0 0.0 0.00 0.00 -89.99
8 0.00 0.00 0.00 0.92 0.09 125000. 10001 0.0 0.0 0.00 0.00 -89.99
9 0.00 0.00 0.00 0.59 0.82 125000. 10001 0.0 0.0 0.00 0.00 -89.99
10 0.00 0.00 0.00 0.30 0.09 125000. 10001 0.0 0.0 0.00 0.00 -89.99
C Total particles, overall phase residual, 10 0.000000

xzhang2005 Mon, 2014-09-22 15:45

You can see that after a global search, all angles and centers are close to zero, and phase residuals are zero too.

Is it ok to set all sefocus to zero? or did this cause the problem? Thanks.

Xing

xzhang2005 Mon, 2014-09-22 19:07

these images were collected with a phase plate, so the correct value would be around zero, but I could set some value such as 2um. I will report the results later. Thanks.

Qianmin Wang Fri, 2015-10-30 10:37

I am testing my phase plate data set with Frealign refinement without doing CTF correction. I set the amplitude contrast to -1.0. However, it doesn't work and the job crashed. I also tried to put the amplitude contrast back to 0.07 and the defocus 0,0. It does work. What is the correct way to run frealign without doing CTF correction?

niko Fri, 2015-10-30 13:27

Can you please give some more detail of how Frealign crashes? Is there a message in the log file? Please remember that if you turn off the CTF, you should not set all defocus values to 0. Set them to something sensible, for example 10000. Also, without CTF correction the contrast in the particle images must be positive, i.e. the protein must be light on a dark background (this is the opposite to what is normally used when CTF correction is switched on). Finally, make sure that the keyword image_contrast is set to N so that Frealign does not try to invert the contrast.

Qianmin Wang Mon, 2015-11-02 04:29

Thanks for your suggestion. It is good to know that the image_contrast should be set to N even the contrast of the particles image is positive. I was doing the refinement by giving the image_contrasct P. I did the following test with 1000 particles:

Trial 1: I tried to switch off the CTF by put the amplitude contrast to -1.0 and defocus values to 0,0,0, the image_contrast P. the job got crashed and the log file shows follow information:
Job 19930 crashed Mon Nov 2 09:32:30 CET 2015
Logfile /fs/fs03/lv16/pool/pool-mizuno-group3/Qianmin/20151016-8mer-gtp-titan2/frealign/test/scratch/oct_mult_refine_n_r1.log_1_7
Final lines:

All rights reserved.
Use is subject to Janelia Farm Research Campus Software Copyright 1.1
license terms ( http://license.janelia.org/license/jfrc_copyright_1_1.html )

CFORM,IFLAG,FMAG,FDEF,FASTIG,FPART,IEWALD,
FBEAUT,FFILT,FBFACT,FMATCH,IFSC,FDUMP,IMEM,INTERP?
M 3 F F F F 0 F F F F 0 F 3 1
PGFIO-F-231/list-directed read/internal file/error on data conversion.
In source file card2.f, at line number 34
Terminating...
frealign.log lines 1-31/31 (END)

Trial 2: Amplitude contrast to -1.0 and defocus values to 0,0,0, the image_contrast N. It refinement worked and no error. So the error information from the trial 1 results from the image_contrast.

As you mentioned that I should give defocus values like 10000, I will change the defocus instead of using 0 and do the final run.

Andy Thu, 2015-07-02 21:04

Hi All,

Following this old thread,

I need to refine datasets, collected with phase plates and 'deliberately' at focus. I already have a density map, computed at a reasonable resolution with other packages (I can avoid ctf correction), but want to try to improve the resolution. In this case, we really donot know what value can be used for defocus.

I am not sure if just using any value like -1.5 µm or -2.0µm arbitrarily, will solve the problem.

What will be an appropriate approach to this issue.

Thanks

Andy

niko Fri, 2015-07-03 00:52

If you really do not know what the effective defocus is you will probably be limited to the resolution where your first CTF zero occurs. If you would like to try switching off the CTF correction in Frealign you can follow this thread. If you do have an idea what the effective defocus is and would like to implement a phase plate CTF to go to higher resolution you can edit the subroutine ctf.f.