Skip to Content

parameter file error

9 replies [Last post]
chaoqi
Offline
Joined: 27 Jul 2017

Hi everyone,

I am a frealign new user. I use Region did 2d classification and use Select in relion2 selected the good particles. Then I use the script on the website converted the star file to par file. But after I run frealign, it still shows the error about parameter file.

Starting refinement...
ERROR: Something wrong with parameter file.
If input file contained film#, defocus 1, defocus 2, astig. angle,
please make sure that numbers are separated by spaces, not commas.

Does anyone have suggestions about it? Thank you very much!!!

Best regards,
Chao

niko
Offline
Joined: 5 May 2009
Please consult the paper on

Please consult the paper on Frealign from 2016 (Frealign: An exploratory tool for single-particle cryo-EM. Methods in Enzymology. 579:191-226). It has examples of what the parameter should look like.

chaoqi
Offline
Joined: 27 Jul 2017
Hi Niko, I prepared the

Hi Niko,

I prepared the parameter file like this:

1 24990.2 25486.4 2.41
1 24990.2 25486.4 2.41
1 24990.2 25486.4 2.41
1 24990.2 25486.4 2.41
1 24990.2 25486.4 2.41
1 24990.2 25486.4 2.41
1 24990.2 25486.4 2.41

I use the constant number for the particle number.No commas,just spaces.
But there is still an error message:
Starting refinement...
ERROR: Something wrong with parameter file.
If input file contained film#, defocus 1, defocus 2, astig. angle,
please make sure that numbers are separated by spaces, not commas.
Cycle 2: refining particles 1 to 385, class 1 Tue Aug 1 17:19:51 CEST 2017
Cycle 2: refining particles 386 to 770, class 1 Tue Aug 1 17:19:51 CEST 2017
Cycle 2: refining particles 771 to 1155, class 1 Tue Aug 1 17:19:51 CEST 2017
Cycle 2: refining particles 1156 to 1540, class 1 Tue Aug 1 17:19:51 CEST 2017

Is the defocusangle in Relion the same parameter with astigmatic angle here? If there are different, I didn't find the astigmatic angle in my star file. Do you think the error come from this parameter?

Thank you very much.

Best regards,
Chao

niko
Offline
Joined: 5 May 2009
The parameter file you

The parameter file you prepared should have been replaced withe a reformatted parameter file that has more columns. Could you most the first line of the reformatted parameter file here so I can see what it looks like?

chaoqi
Offline
Joined: 27 Jul 2017
Hi Niko, These are the first

Hi Niko,

These are the first several lines after the script transform.

1 -103.45 18.29 135.17 -12.03 -4.63 10000 1 24990.2 25486.4 2.41 100.00 -500 1.0000 20.00 0.00
2 -63.73 93.82 -1.92 -4.63 1.72 10000 1 24990.2 25486.4 2.41 100.00 -500 1.0000 20.00 0.00
3 -63.17 115.43 -147.93 2.78 8.07 10000 1 24990.2 25486.4 2.41 100.00 -500 1.0000 20.00 0.00
4 -27.27 109.55 93.37 8.07 -3.57 10000 1 24990.2 25486.4 2.41 100.00 -500 1.0000 20.00 0.00
5 161.17 108.51 3.68 -7.80 14.42 10000 1 24990.2 25486.4 2.41 100.00 -500 1.0000 20.00 0.00
6 -124.29 120.84 149.68 -0.39 1.72 10000 1 24990.2 25486.4 2.41 100.00 -500 1.0000 20.00 0.00

I did 3d classification with class=1 using relion. Someone said that it is better do one round of 3d classification or refinement before move to frealign. I don't know it is true or not.

Best regards,
Chao

niko
Offline
Joined: 5 May 2009
It looks like this parameter

It looks like this parameter file is not the file that was converted from the earlier file you showed, containing four numbers on each line. Where did this file come from?

chaoqi
Offline
Joined: 27 Jul 2017
Dear Niko, Sorry for wasting

Dear Niko,

Sorry for wasting your time. I think I found the reason. My path is wrong.

Best regards,
Chao

chaoqi
Offline
Joined: 27 Jul 2017
Hi Niko, Because it shows

Hi Niko,

Because it shows error at the first step, I didn't get the reformatted parameter file.

Best regards,
Chao

chaoqi
Offline
Joined: 27 Jul 2017
Dear Niko, The new parameter

Dear Niko,

The new parameter file is produced directly with the script on the website (relion to frealign). It is not working. Then I modified the file, just take out the particle number and defucus values. But unfortunately, it is still not working.

Best regards,
Chao

niko
Offline
Joined: 5 May 2009
Could you test your

Could you test your installation with the test dataset? If this works you can compare your parameter file that you converted from Relion to the one generated by refining the test dataset.