I would recommend to use the option to ignore the CTF until the first
peak, instead of only phase flipping. That way, you will still correct all
high-frequency phases and amplitudes, but you will ignore the lowest
frequencies, which probably cause this problem.
HTH,
Sjors
> Hi Christian,
>
> The ususal --only_flip_phases option and limiting the resolution to 10-ang
> or so should help.
>
> Ludo
>
> On Fri, Sep 4, 2015 at 3:08 PM, Christian Benda <[log in to unmask]>
> wrote:
>
>> Dear all
>>
>> I am facing a problem with 2D classification in relion (1.3 or 1.4).
>>
>> I have a set of automatically picked particles (10k to 50k) wich
>> contains
>> many "good" but also bad (e.g. from carbon or the hole edge) particles.
>> During 2D classification all particles end up in a single or very few
>> bad
>> classes and all other classes remain empty (black) - no matter how many
>> classes I am asking for.
>>
>> Has anyone faced a similar situation and knows what knob to turn or
>> which
>> settings to play with? I tried T=1 or 2 so far without improvement.
>>
>> Many thanks in advance
>>
>> Christian
>>
>
--
Sjors Scheres
MRC Laboratory of Molecular Biology
Francis Crick Avenue, Cambridge Biomedical Campus
Cambridge CB2 0QH, U.K.
tel: +44 (0)1223 267061
http://www2.mrc-lmb.cam.ac.uk/groups/scheres
|