EER format

[Jason Ja]

Hi all,

Hope this message finds you well.

I’ve been using eman2 to check the headers of EER movies, and I noticed that the parameters displayed differ from those set in EPU. I was wondering if anyone might have insights into what could be causing this discrepancy?

Many thanks in advance for your help.

Best regards,
Jason

[Steve Ludtke]

Hi Jason,

you'll need to specify which header values you're referring to. EER is basically a TIFF file, meaning the header can include whatever tags the company decides to add, and the list can grow with time. Unless someone notices, we don't know that we need to expand the parameters. However, that's different from reading a value and the value being incorrect. If there are missing header values you need, let us know what they are and we can quickly add them to the reader. If there are incorrect values, that's more serious, and we definitely want to know.

cheers

--
--
----------------------------------------------------------------------------------------------
You received this message because you are subscribed to the Google
Groups "EMAN2" group.
To post to this group, send email to em...@googlegroups.com
To unsubscribe from this group, send email to eman2+un...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/eman2

---
You received this message because you are subscribed to the Google Groups "EMAN2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eman2+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/eman2/335dbe55-34ff-4be5-8bda-99d17a2e235dn%40googlegroups.com.

[Jason Ja]

Hi  Steve,

My data collection paremeter is :

Tne eman2 readed parameters is :

e2iminfo.py FoilHole_7708159_Data_7705510_7705512_20240509_055637_EER.eer --header

FoilHole_7708159_Data_7705510_7705512_20240509_055637_EER.eer       1512 images in unknown format (swap)  4096 x 4096

0.           

                EER.acquisition_id: 20240509.055637.472

                EER.camera_name: EF-Falcon

                EER.commercial_name: Falcon 4i

                EER.compression: 65001

                EER.eer_gain_reference: ImagesForProcessing/EF-Falcon/300kV/20240415_090525_EER_GainReference.gain

                EER.exposure_time: 4.9156139999999997

                EER.mean_dose_rate: 9.1467210223587117

                EER.number_of_frames: 1512

                EER.sensor_image_height: 4096

                EER.sensor_image_width: 4096

                EER.sensor_pixel_size.height: 9.56559346e-11

                EER.sensor_pixel_size.width: 9.56559346e-11

                EER.serial_number: 21-24-A1F-OX2

                EER.timestamp: 2024-05-09T05:56:37.932843+08:00

                EER.total_dose: 44.932965180543434

                HostEndian: little

                ImageEndian: big

                apix_x: 0.9565593600273132

                apix_y: 0.9565593600273132

                apix_z: 0.9565593600273132

                changecount: 0

                datatype: 7

                is_complex: 0

                is_complex_ri: 1

                is_complex_x: 0

                nx: 4096

                ny: 4096

                nz: 1

                source_n: 0

                source_path: FoilHole_7708159_Data_7705510_7705512_20240509_055637_EER.eer

What caused this ?

Best,

Jasion

[Steve Ludtke]

Hi Jason,

So, two discrepancies. Å/pix is different and dose is different. My suspicion is that if you open the EER file on the microscope, you will see the same values EMAN2 shows. EMAN2 should not be altering any of the header values when it reads the file.  So, why are these numbers different?  See:

https://assets.thermofisher.com/TFS-Assets/MSD/Support-Files/eer-file-format.pdf

- Dose:

I assume that in your table 60 e-/Å^2 is the specimen exposure. The value stored in the EER file is "Average amount of electrons per pixel in the entire acquisition". This incorporates 2 differences:

- First, the EER value is per pixel, not per Å^2. You're close to 1 Å/pixel, but not quite the same.

- Second, this is the dose seen by the detector. Some electrons will be lost when the beam passes through the specimen. ie- this is recorded dose, not incident dose.

- A/pix:

- This is a bit trickier. First, it's important to understand that the magnification setting on TEMs is only approximate, and needs to be calibrated to produce an accurate value. Its can also vary over time, typically by 2-4%. Even if you think you've done a good job calibrating the instrument, it still won't be perfect. This is why when doing a near-atomic resolution single particle structure it is usually necessary to do a final recalibration of the A/pix based on the model. This is also why it can be very tricky to combine two data sets collected on different microscopes or at significantly different times on the same microscope when targeting very high resolution. 

If you look at your setting, 14 um/pixel / 130,000 = 1.077 Å/pix, so neither the value in your table or the value in the EER file agree well with the standard computed value. Do you have a Selectris energy filter on this instrument?  I don't have one, so not sure, but it is possible that there is an additional magnification modifier in this setup. Otherwise this would be a 15% calibration deviation which seems quite large. 

As to the difference between the 0.925 in your table and the 0.9565 in the file, the EER specifications don't specify how this value is computed. It's a little surprising that it is different than the value you see in the collection software, but I'm pretty confident that the value EMAN2 shows you is the actual value stored in the file. 

On Aug 15, 2025, at 5:47 AM, Jason Ja <jj53...@gmail.com> wrote:

Hi  Steve,

My data collection paremeter is :

To view this discussion visit https://groups.google.com/d/msgid/eman2/8954c25d-444d-4a05-b737-0d5e3045ed84n%40googlegroups.com.
<2025-08-15 18-45-04.png>

[Jason Ja]

Hi Steve,

Thank you so much for your assistance. I have one final question:

If I need to process those images using other software such as cryoDRAGON, or RELION, which parameters should I use—the EMAN2 header readed or the EPU settings?

Best regards,
Jason

[Steve Ludtke]

For the dose, you generally want the incident dose on the specimen in e-/A^2, not the dose in e-/pixel.  

For the A/pix, which is obviously a critical parameter, you need to speak with the microscope operator. Most cryoEM labs will ensure that they have a reasonably accurate calibrated A/pix value for each magnification which should be within a percent or so of the correct value. Even a 1-2% error can cause some minor issues which have to be dealt with when model-building at high resolution. The variances you're seeing here are _much_ larger, and you really need to talk to the microscope operator to find out what the correct value is. You might also ask them why you're seeing such different values. While it is generally possible to correct a ~1% error by rescaling the map when model building, if you're off by 15%, then CTF determination/correction may not work properly at high resolution.

To view this discussion visit https://groups.google.com/d/msgid/eman2/f9cffbaf-d661-4ce6-b145-86e049450391n%40googlegroups.com.