Hi,
For Class3D, it is more important to visually inspect the maps
in both slices and 3D for better features (conformation, presence of
cofactor, lack of artifacts and blurriness) than comparing numbers.
(But this is also the reason why full automation of SPA is difficult.
Although programs for automatic selection of 2D class averages are emerging,
automatic selection of 3D classes is more difficult)
Best regards,
Takanori Nakane
On 2020/09/12 3:06, 史 沙沙 wrote:
> Thanks for your help!
>
> Sorry if this sounds a stupid question to ask. For 3D classification, how do you know whether your final result is the best? Here in my case, I just did another 3D classification and found the resolution decreased. And at what resolution do you think is good enough to go for the auto-refine?
>
> Shasha
>
>> On Sep 11, 2020, at 4:44 PM, Takanori Nakane <[log in to unmask]> wrote:
>>
>> Hi,
>>
>> Class3D does not use half-set separation. Thus, the resolution estimates in
>> Class3D are unreliable.
>>
>>> I used initial low-pass filter of 50A.
>>
>> For small proteins, sometimes you need to change this to 20 A or even 15 A.
>> (But since your final resolution is only around ~ 10 A, I don't recommend
>> going too close to it. 20 A is probably OK. 15 A is dangerous)
>>
>> Best regards,
>>
>> Takanori Nakane
>>
>> On 2020/09/11 22:38, 史 沙沙 wrote:
>>> Hi all, After I get a good 3D class with a resolution of 10A, I re-extracted the selected particle to with less down-scaling and did the auto-refinejob. My particle is small with the diameter 100/160A. I used initial low-pass filter of 50A. After 3D auto-refine, the final resolution without maskincreased to 12. The map from refine job is much worse than the 3D classification result. I though after 3D auto-refine, we should get a betterresolution, am I right? Any comments or suggestion are highly welcome.
>>> Thanks
>>> Shasha
>>> ------------------------------------------------------------------------------------------------------------------------------------------------------
>>> To unsubscribe from the CCPEM list, click the following link:
>>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCPEM&A=1
>>
>
>
> ########################################################################
>
> To unsubscribe from the CCPEM list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCPEM&A=1
>
> This message was issued to members of www.jiscmail.ac.uk/CCPEM, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
>
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCPEM&A=1
This message was issued to members of www.jiscmail.ac.uk/CCPEM, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|