[3dem] Tilt axis angle

[Ricardo Righetto]

Hi Sylvain,

> Or maybe this is a flip and then the orientation would be ok, but this
would change the handedness.
Yes, this is what I think might be happening here. This "off by just 10
degrees" is misleading, it's actually the expected 5 degrees deviation from
vertical, but going over to the other side of the axis. Your tilt series
images are probably being flipped somewhere in your pipeline (perhaps at
the detector itself).
Which is not necessarily a problem, as long as the tilt axis and tilt
angles are specified accordingly.

Best wishes,

--
Ricardo Diogo Righetto


Em seg., 18 de mar. de 2024 às 02:57, Sylvain Trepout <
sylvain.trepout at monash.edu> escreveu:

> Hi Zongli and Ricardo, and tomo enthusiasts,
>
> Same here, on a Krios G4 with Falcon 4i and Tomo software version
> 5.16.0.6389, the orientation of the tilt axis is not reported correctly.
> TiltAxisAngle = -174.72
> RotationAngle = 84.72
>
> But the correct tilt axis angle found by IMOD is 95.07
> It does not fit the formula you gave Ricardo though... off by just 10
> degrees.
> Or maybe this is a flip and then the orientation would be ok, but this
> would change the handedness.
>
> For comparison, on the Krios G1 (with a K3 and an older version of Tomo
> software)
> TiltAxisAngle = -174.47
> RotationAngle = -84.72
> The correct tilt axis orientation (as found by IMOD) is around 87 (this
> fits your formula, Ricardo).
>
> Best wishes,
> Sylvain
>
> PhD
> Research Fellow - Microscopy Australia
>
> *Monash University*
> Ramaciotti Centre for Cryo-EM
> G92, 15 Innovation Walk, Clayton Campus
> Victoria, 3800 Australia
>
> E: sylvain.trepout at monash.edu
>
>
> On Sun, 17 Mar 2024 at 07:53, Zongli Li <zongli.li at gmail.com> wrote:
>
>> Dear Ricardo,
>>
>> Thank you so much! This makes a lot of sense.
>> Best wishes,
>>
>> Zongli
>>
>> On Sat, Mar 16, 2024 at 12:37 PM Ricardo Righetto <
>> ricardorighetto at gmail.com> wrote:
>>
>>> Dear Zongli,
>>>
>>> Thanks for bringing this up. As far as I know this is a bug (?) in
>>> Tomo5. This is how you convert the TiltAxisAngle from the Tomo5 mdoc to
>>> match the IMOD/SerialEM convention (followed by RELION and other packages):
>>>
>>> TiltAxisAngle_SerialEM = TiltAxisAngle_Tomo5 * (-1) - 90
>>>
>>> Thanks to Grigory Sharov for reporting this conversion previously on the
>>> EMAN2 mailing list:
>>> https://urldefense.com/v3/__https://groups.google.com/g/eman2/c/piux0GdX_7M/m/ScQNh9TNAQAJ__;!!Mih3wA!Cm7XcxF6VqlPRi_nuhuhA5UJxZ1CPlAv8FHSp5bPNetfDBaUg5hUU07A7CuZSCG0fmFna5-2cySw_u-xToQzMzQ2IizT$ 
>>> <https://urldefense.com/v3/__https://groups.google.com/g/eman2/c/piux0GdX_7M/m/ScQNh9TNAQAJ__;!!Mih3wA!Go-c8e-QwJVBsU0qttyzgSDv-knvzqN6KJbTdN11SKI-RKgQ9dktWrwtqL_J9HC10IqJx5KxDD2MNvPZcrYB$>
>>>
>>> TiltAxisAngle and RotationAngle should always be 90 degrees apart in any
>>> mdoc:
>>> https://urldefense.com/v3/__https://bio3d.colorado.edu/SerialEM/hlp/html/about_formats.htm__;!!Mih3wA!Cm7XcxF6VqlPRi_nuhuhA5UJxZ1CPlAv8FHSp5bPNetfDBaUg5hUU07A7CuZSCG0fmFna5-2cySw_u-xToQzM_tT6y4J$ 
>>> <https://urldefense.com/v3/__https://bio3d.colorado.edu/SerialEM/hlp/html/about_formats.htm__;!!Mih3wA!Go-c8e-QwJVBsU0qttyzgSDv-knvzqN6KJbTdN11SKI-RKgQ9dktWrwtqL_J9HC10IqJx5KxDD2MNiJtThea$>
>>> That's why you got the "correct" result when using the value from
>>> RotationAngle as the tilt axis. Beware though that you may have inverted
>>> handedness in your reconstructions. In my experience, the conversion above
>>> should give the correct results.
>>>
>>> Would be great if others can comment if this is indeed the case, and why
>>> this bug persists for so long...
>>>
>>> Best wishes,
>>>
>>> --
>>> Ricardo Diogo Righetto
>>>
>>>
>>> Em sex., 15 de mar. de 2024 às 21:33, Zongli Li <zongli.li at gmail.com>
>>> escreveu:
>>>
>>>> Dear colleagues,
>>>>
>>>> We recently collected some tilt series data on vitrified ribosome
>>>> sample using ThermoFisher Scientifc's Tomo5 software. The data were
>>>> collected on Falcon 4i. When trying to process the data using the new
>>>> version of relion tomo workflow (relion 5.0-beta), we are wondering what
>>>> parameter we should use from the dmoc file for tilt axis angle
>>>> (TiltAxisAngle or RotationAngle). If the TiltAxisAngle is used in our test
>>>> run, the reconstructed tomogram is nonsense, but the reconstructed tomogram
>>>> looks okay if RotationAngle is used. Any suggestions or explanations on
>>>> this would be highly appreciated.
>>>>
>>>> Best,
>>>>
>>>> Zongli
>>>> _______________________________________________
>>>> 3dem mailing list
>>>> 3dem at ncmir.ucsd.edu
>>>> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
>>>>
>>> _______________________________________________
>> 3dem mailing list
>> 3dem at ncmir.ucsd.edu
>> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20240318/f737468d/attachment.html>