search

LBEM-CH / focus

High Throughput Electron Microscopy Image Processing Software
http://www.focus-em.org
GNU General Public License v2.0
35 stars 14 forks source link

2D classification script error and explanation of numerical results #209

Open ajrzepiela opened 6 years ago

ajrzepiela commented 6 years ago

Dear Focus developers,

I have 2 issues with which I would like to get some help:

  1. For the Relion 2D classification script (in the Datasets tab) I get error "Default_CS :Undefined variable" . Is this script already functional ? Any tips how to fix the error and make the script work?

  2. For the result column Dse/frm[e/A2] in the Library tab we obtain somehow too high numbers (e.g around 36). Can I get a bit more detailed explanation from which input parameters this value is calculated? Playing with the parameters ourselves so far we cannot figure out how to make the value in the range as we think it should be.

Thank you,

Best regards, Andrzej

rdrighetto commented 6 years ago

Hi Andrzej,

  1. You need to set the value for "Default_CS" in Preferences >> Microscope. The script is functional but still experimental. It assumes that you have picked particles, extracted and normalize them during the processing of each movie previously. You can enable/adjust that from the scripts under "Particles" (specifically, scripts "Pick Particles" and "Normalize Particles" e.g. if you double-click one of the movies in your library. The idea is that you should leave the Relion 2D classification script running continuously (like every hour or so) with increasing number of picked particles from your session. Or you can just run it manually from time to time to assess the quality of your particles.

  2. A value of 36 would look normal for a full movie (Total Dse [e/A2]), but for Dse/frm it is indeed strange. By default, this is measured from the counts in the image and then conversion factors can optionally be applied. Please check the parameters under "Electron Dose" in the projects Tab, or in your script Prepare >> "Import Movie from SerialEM" (or "Import Movie from EPU" depending on what you use). As you will see there, you can also impose a value manually if you prefer.

Please let us know if you need further assistance.

Best wishes,

-- Ricardo Diogo Righetto

2018-01-17 16:09 GMT+01:00 arzepiela notifications@github.com:

Dear Focus developers,

I have 2 issues with which I would like to get some help:

1.

For the Relion 2D classification script (in the Datasets tab) I get error "Default_CS :Undefined variable" . Is this script already functional ? Any tips how to fix the error and make the script work? 2.

For the result column Dse/frm[e/A2] in the Library tab we obtain somehow too high numbers (e.g around 36). Can I get a bit more detailed explanation from which input parameters this value is calculated? Playing with the parameters ourselves so far we cannot figure out how to make the value in the range as we think it should be.

Thank you,

Best regards, Andrzej

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/C-CINA/focus/issues/209, or mute the thread https://github.com/notifications/unsubscribe-auth/AHVvf8dDEhlV8aBFOdJbexlRuRRpwIz4ks5tLg0hgaJpZM4RhfIW .

ajrzepiela commented 6 years ago

Hey Ricardo,

1) Default Microscope CS is set in the Preferences tab to 2.7, high tension to 300. The error is there with these variables set. I can try to debug it myself but it would be great if you could take a look.

2) Here I am step further. Setting electron dose to manual and setting all the parameters below in the Electron Dose section gives the effect on Dse/frm. But the total dose is not calculated (empty field left). What this can be?

Thanks,

Andrzej

On 18 Jan 2018, at 10:32, Ricardo Righetto notifications@github.com wrote:

Hi Andrzej,

  1. You need to set the value for "Default_CS" in Preferences >> Microscope. The script is functional but still experimental. It assumes that you have picked particles, extracted and normalize them during the processing of each movie previously. You can enable/adjust that from the scripts under "Particles" (specifically, scripts "Pick Particles" and "Normalize Particles" e.g. if you double-click one of the movies in your library. The idea is that you should leave the Relion 2D classification script running continuously (like every hour or so) with increasing number of picked particles from your session. Or you can just run it manually from time to time to assess the quality of your particles.

  2. A value of 36 would look normal for a full movie (Total Dse [e/A2]), but for Dse/frm it is indeed strange. By default, this is measured from the counts in the image and then conversion factors can optionally be applied. Please check the parameters under "Electron Dose" in the projects Tab, or in your script Prepare >> "Import Movie from SerialEM" (or "Import Movie from EPU" depending on what you use). As you will see there, you can also impose a value manually if you prefer.

Please let us know if you need further assistance.

Best wishes,

-- Ricardo Diogo Righetto

2018-01-17 16:09 GMT+01:00 arzepiela notifications@github.com:

Dear Focus developers,

I have 2 issues with which I would like to get some help:

1.

For the Relion 2D classification script (in the Datasets tab) I get error "Default_CS :Undefined variable" . Is this script already functional ? Any tips how to fix the error and make the script work? 2.

For the result column Dse/frm[e/A2] in the Library tab we obtain somehow too high numbers (e.g around 36). Can I get a bit more detailed explanation from which input parameters this value is calculated? Playing with the parameters ourselves so far we cannot figure out how to make the value in the range as we think it should be.

Thank you,

Best regards, Andrzej

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/C-CINA/focus/issues/209, or mute the thread https://github.com/notifications/unsubscribe-auth/AHVvf8dDEhlV8aBFOdJbexlRuRRpwIz4ks5tLg0hgaJpZM4RhfIW .

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/C-CINA/focus/issues/209#issuecomment-358589288, or mute the thread https://github.com/notifications/unsubscribe-auth/ASnbwrWzjMkRdzoKimMeaewldmE4KPWnks5tLw-rgaJpZM4RhfIW.

rdrighetto commented 6 years ago

Hi Andrzej,

1) There was indeed a bug in the script that would prevent it from reading Default_CS and Default_KV in some circunstances. This has been corrected and updated on GitHub, please try again.

2) Total dose should be correctly calculated in any case. Not sure what is going on. I just tested for the "Import Movie from SerialEM" with manual and automatic frame dose definition, and "Import Movie from EPU" (which reads the dose from a .xml file). Could you please tell which script are you using exactly?

Best wishes,

-- Ricardo Diogo Righetto

2018-01-19 15:47 GMT+01:00 arzepiela notifications@github.com:

Hey Ricardo,

1) Default Microscope CS is set in the Preferences tab to 2.7, high tension to 300. The error is there with these variables set. I can try to debug it myself but it would be great if you could take a look.

2) Here I am step further. Setting electron dose to manual and setting all the parameters below in the Electron Dose section gives the effect on Dse/frm. But the total dose is not calculated (empty field left). What this can be?

Thanks,

Andrzej

On 18 Jan 2018, at 10:32, Ricardo Righetto notifications@github.com wrote:

Hi Andrzej,

  1. You need to set the value for "Default_CS" in Preferences >> Microscope. The script is functional but still experimental. It assumes that you have picked particles, extracted and normalize them during the processing of each movie previously. You can enable/adjust that from the scripts under "Particles" (specifically, scripts "Pick Particles" and "Normalize Particles" e.g. if you double-click one of the movies in your library. The idea is that you should leave the Relion 2D classification script running continuously (like every hour or so) with increasing number of picked particles from your session. Or you can just run it manually from time to time to assess the quality of your particles.

  2. A value of 36 would look normal for a full movie (Total Dse [e/A2]), but for Dse/frm it is indeed strange. By default, this is measured from the counts in the image and then conversion factors can optionally be applied. Please check the parameters under "Electron Dose" in the projects Tab, or in your script Prepare >> "Import Movie from SerialEM" (or "Import Movie from EPU" depending on what you use). As you will see there, you can also impose a value manually if you prefer.

Please let us know if you need further assistance.

Best wishes,

-- Ricardo Diogo Righetto

2018-01-17 16:09 GMT+01:00 arzepiela notifications@github.com:

Dear Focus developers,

I have 2 issues with which I would like to get some help:

1.

For the Relion 2D classification script (in the Datasets tab) I get error "Default_CS :Undefined variable" . Is this script already functional ? Any tips how to fix the error and make the script work? 2.

For the result column Dse/frm[e/A2] in the Library tab we obtain somehow too high numbers (e.g around 36). Can I get a bit more detailed explanation from which input parameters this value is calculated? Playing with the parameters ourselves so far we cannot figure out how to make the value in the range as we think it should be.

Thank you,

Best regards, Andrzej

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/C-CINA/focus/issues/209, or mute the thread https://github.com/notifications/unsubscribe-auth/ AHVvf8dDEhlV8aBFOdJbexlRuRRpwIz4ks5tLg0hgaJpZM4RhfIW .

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub < https://github.com/C-CINA/focus/issues/209#issuecomment-358589288>, or mute the thread https://github.com/notifications/unsubscribe-auth/ ASnbwrWzjMkRdzoKimMeaewldmE4KPWnks5tLw-rgaJpZM4RhfIW.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/C-CINA/focus/issues/209#issuecomment-358985911, or mute the thread https://github.com/notifications/unsubscribe-auth/AHVvf5c6x_koPMSCnEnLOPpdhxbxdx-Wks5tMKsWgaJpZM4RhfIW .